aw

W
.P

.A.

u-

11% m» '” a

“1.! w- ’7'...

-‘J’é‘;

‘.. ‘-
.' ‘E‘

{amum‘

.5—

"1‘.
3‘

3h .

2n-

rho-or

”a

34'

d'.

 

xix;
; it 1'”

h
..
H

a
:0

\
".
I
.r

.

'3

.. -‘-
a. .3. .. ‘ ,,.

u
:1»:

fig:

-~

 

 

 

 

 

 

 

 

v

9.».an
1‘0 aA-u .m-

7hr" A
-. .. '1’»?-
i: mgr?"

. ...,. ..:..

“WP...“ ..
- x

m “-

my ‘l
‘

 

L
E UU‘MVER
l

M \Hll‘llll

ICHl SITY LIBRARIES

ll l will lllll

0
ll 1‘
ll
3 1293 01026 9 97

 

l

 

 

 

 

l
l
l

 

 

 

 

Jl

 

 

This is to certify that the
thesis entitled
NEW EMPLOYEE ADAPTATION: AN EXPLORATION OF

INFORMATION-SEEKING, ITS ANTECEDENTS, AND OUTCOMES

presented by

Eleanor Marie Smith

has been accepted towards fulfillment
of the requirements for

M. A. degree in Psychology

MW/

ajor’pér ss ‘—
e 0% gal. #23 / fl

0—7 639

Dat

 

MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE II RETURN BOX to romovo thlo chockout from your rocord.
To AVOID FINES rotum on or bdoro doto duo.

 

4

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

——lI———
=ffi’.

MSUIoAnNflmotlvoAcfloNEmolOppoMylnofltwon

 

 

 

 

 

 

 

NEW EMPLOYEE ADAPTATION: AN EXPLORATION OF
INFORMATION-SEEKING, ITS ANTECEDENT S, AND OUTCOMES

By

Eleanor Marie Smith

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology

1993

ABSTRACT

NEW EMPLOYEE ADAPTATION: AN EXPLORATION OF
INFORMATION-SEEKING, ITS ANTECEDENT S, AND OUTCOMES

By

Eleanor Marie Smith

Recent socialization theories describe the newcomer as a proactive participant
in his/her adaptation to the new job, seeking out information to attain personal goals.
This study identified factors that may affect the strategies used to acquire functional
and evaluative information about the job. Relationships between information-seeking
strategies and outcomes of socialization were also examined.

Hypotheses were tested using a sample of 253 individuals with job tenure of
approximately one year or less. At least one coworker for each new employee also
provided ratings for the socialization outcomes. Results indicated that self-efficacy
and several job characteristics predicted the use of infomiation-seeking strategies.
Furthermore, supervisor and coworker information-sharing was positively related to the
frequency of newcomer information-seeking. Finally, several information-seeking
strategies were related to self ratings of mastery and innovation. When coworker
ratings of socialization outcomes were used, few information-seeking strategies

affected socialization outcomes.

ACKNOWLEDGEMENTS

This master’s thesis has been a rewarding learning experience for a newcomer
to organizational research. My adaptation to the role of a researcher mirrors some of
the same processes described in this thesis. While I relied to a large extent on my
own efforts and my developing ability to adapt to unforeseen circumstances, I was also
fortunate for the opportunities to benefit from the expertise of several individuals.

I would like to thank my committee chairperson, Steve Kozlowski, for his help
and guidance through every step of this research project. He provided me with
assistance when needed, but also allowed me the latitude and responsibility for
ultimately determining the direction and scope of my thesis. I would also like to
thank the other members of my committee, Georgia Chao and Neal Schmitt. Their
advice and encouragement made this thesis a better quality research product.

This project would not have been possible without the funding provided by
Philip Gardner of the Collegiate Employment Research Institute. I appreciate the
administrative support he provided in terms of materials, mailing costs, and assistance
in entering data, as well as his kind words of advice and support.

Finally, I am grateful to Sean Tyler and the rest of my family for the love and

encouragement they provided throughout this process.

TABLE OF CONTENTS

Page

LIST OF TABLES ................................................ viii
LIST OF FIGURES ............................................... xi
THEORIES OF NEWCOMER SOCIALIZATION ........................... 1
Stage Models ............................................... 1
Socialization Tactics .......................................... 3

An Interactionist Perspective .................................... 6

A CONCEPTUAL MODEL OF NEWCOMER LEARNING AND ADAPTATION . . . 10

Socialization Outcomes ....................................... lO
Newcomer Information-Seeking ................................. 13
Information content .................................... l4
Information-seeldng strategies ............................. 15
Strategy choice ....................................... 17
Research on Information-Seeking ................................ l8
Differential Use of Information-Seeking Strategies .................... 22
Individual Difference Factors ................................... 26
Self-Efficacy ....................................... 28
Job Characteristics .......................................... 32
Interdependence ....................................... 32
Complexity .......................................... 38
Discretion ........................................... 47

iv

Workgroup Relations ......................................... 53

Supervisor and coworker information-sharing .................. 54
Leader-member exchange relationships ....................... 55
Team-member exchange relationships ........................ 60
Outcomes of Newcomer Information-Seeking ........................ 64
METHOD ...................................................... 68
Sample and Procedure ........................................ 68
Measures ................................................. 73
Job status ........................................... 73
Time on the job ....................................... 73
Inquiry of supervisor - functional information .................. 74
Inquiry of supervisor — evaluative information .................. 74
Inquiry of coworkers - functional information .................. 74
Inquiry of coworkers - evaluative information .................. 74
Observation - functional information ......................... 75
Observation - evaluative information ........................ 75
Experimentation - functional information ..................... 75
Experimentation - evaluative information ..................... 75
Self-efficacy ......................................... 80
Interdependence ....................................... 81
Complexity .......................................... 81
Discretion ........................................... 82
Supervisor functional information-sharing ..................... 84
Supervisor evaluative infonnation-sharing ..................... 84
Coworker functional information-sharing ...................... 85
Coworker evaluative information-sharing ..................... 85
Leader-member exchange ................................ 87
Team-member exchange ................................. 87
Mastery ............................................. 88
Innovation ........................................... 90
Workgroup influence ................................... 91
RESULTS ...................................................... 93
Patterns of Information-Seeking by Strategy, Content,
and Time on the Job .................................... 93
Hypothesis 1 ......................................... 93
Hypotheses 2 and 4 ................................... 102
Hypotheses 5 and 7 ................................... 106
Hypothesis 8 ........................................ 107

Hypothesis 9 ........................................ 108

Hypothesis lO ....................................... 109
Hypothesis 11 ....................................... 110

Job Characteristics and Self-Efficacy as Antecedents
to Information-Seeking ................................. 116
Hypotheses 12, 12a, 20, 20a, 28, 28a, 36, and 36a .............. 117
Hypotheses 13, 13a, 21, 21a, 29, 29a, 37, and 37a .............. 119
Hypotheses 14, 14a, 22, 22a, 30, 30a, 38, and 38a .............. 119
Hypotheses 15, 15a, 23, 23a, 31, 31a, 39, and 39a .............. 122
Hypotheses 16, 16a, 24, 24a, 32, 32a, 40, and 40a .............. 122
Hypotheses 17, 17a, 25, 25a, 33, 33a, 41, and 41a .............. 125
Hypotheses 19, 19a, 27, 27a, 35, 35a, 43, and 43a .............. 125
Workgroup Relations and Information-Seeking ...................... 128
Hypotheses 44 and 52 .................................. 129
Hypothesis 45 ....................................... 132
Hypothesis 46 ....................................... 132
Hypothesis 47 ....................................... 132
Hypothesis 48 and 53 .................................. 133
Hypothesis 49 ....................................... 133
Hypothesis 51 ....................................... 134
Information-Seeking and Socialization Outcomes .................... 134
Hypotheses 56-58 ..................................... 135
Hypothesis 59 ....................................... 137
Hypothesis 60 ....................................... 140
Hypothesis 61 ....................................... 143
Hypotheses 62 and 63 .................................. 145
DISCUSSION .................................................. 148
Pattems of Information-Seeking ................................ 148
Antecedent Factors and Information-Seeking ....................... 149
Information-Seeking and Socialization Outcomes .................... 158
Summary and Implications .................................... 164
Directions for Future Research ................................. 165
APPENDIX A: Inquiry of Supervisor - Functional Information Items ........... 170

vi

APPENDIX B: Inquiry of Supervisor - Evaluative Information Items ........... 171
APPENDDI C: Inquiry of Coworkers - Functional Information Items ........... 172
APPENDIX D: Inquiry of Coworkers - Evaluative Information Items ........... 173
APPENDDI E: Observation - Functional Information Items .................. 174
APPENDDI F: Observation - Evaluative Information Items .................. 175
APPENDIX G: Experimentation - Functional Information Items .............. 176
APPENDIX H: Experimentation - Evaluative Information Items .............. 177
APPENDDI I: Self-Efficacy Items ................................... 178
APPENDIX J: Interdependence Items ................................. 179
APPENDDI K: Complexity Items .................................... 180
APPENDIX L: Discretion Items ..................................... 181
APPENDIX M: Supervisor Functional Information-Sharing Items ............. 182
APPENDIX N: Supervisor Evaluative Information-Sharing Items .............. 183
APPENDIX 0: Coworker thctional Infonnation-Sharing Items .............. 184
APPENDIX P: Coworker Evaluative Infonnation-Sharing Items ............... 185
APPENDIX Q: Leader-Member Exchange Items ......................... 186
APPENDIX R: Team-Member Exchange Items .......................... 188
APPENDIX S: Mastery Items ...................................... 190
APPENDIX T: Innovation Items .................................... 191
APPENDIX U: Workgroup Influence Items ............................. 192
APPENDIX V: Study Analyses ..................................... 193
APPENDIX W: Principle Axis Factor Matrix for Observation

and Coworker Sharing of Functional Information ............... 204
LIST OF REFERENCES .......................................... 205

vii

Table 1

Table 2
Table 3
Table 4

Table 5

Table 6
Table 7
Table 8

Table 9

Table 10

Table 11

Table 12

Table 13

Table 14

Table 15

LIST OF TABLES

Page

Discriminant Analysis Results for

Professional/Nonprofessional Job Status ...................... 72
Principal Axis Factor Matrix for Information-Seeking Items ............ 77
Principal Axis Factor Matrix for Self-Efficacy Items ................. 81
Principal Axis Factor Matrix for Job Characteristics Items ............. 83
Principal Axis Factor Matrix for Supervisor and

Coworker Assistance Items ............................... 86
Principal Axis Factor Matrix for Mastery Items ..................... 89
Principal Axis Factor Matrix for Innovation Items ................... 91
Principal Axis Factor Matrix for Workgroup Influence Items ........... 92
Hypothesis 1: 3 (Strategy) x 2 (Content) x

3 (Time on the Job) ANCOVA Results ....................... 95
Three-Way ANCOVA Results:

Cell Means and Standard Deviations ......................... 97
Understanding the Three-Way Interaction: Test of

Simple Interaction Effects at Each Tenure Level ................ 99
Understanding the 3-Way Interaction: Test of Simple

Interaction Effects for Each Information Content Category ........ 101
Hypotheses 2 and 4: AN OVA Results for

Functional Information-Seeking ........................... 103
Hypotheses 2 and 4: ANOVA Results for

Functional Information-Seeking at Each Tenure Level ........... 104
Hypotheses 5 and 7: ANOVA Results for

Evaluative Information-Seeking ........................... 107

viii

Table 16

Table 17

Table 18

Table 19

Table 20

Table 21

Table 22

Table 23

Table 24

Table 25

Table 26

Table 27

Table 28

Table 29

Hypothesis 8: AN OVA Results for Inquiry
of Supervisor for Functional Information ..................... 108

Hypothesis 9: AN OVA Results for Inquiry
of Coworkers for Functional Information .................... 109

Hypothesis 10: AN OVA Results for Inquiry

of Supervisor for Evaluative Information ..................... 111
Hypothesis 11: ANOVA Results for Inquiry

of Coworkers for Evaluative Information .................... 111
Means, Standard Deviations, and Intercorrelations .................. 114

Moderated Hierarchical Regression Results
Dependent Variable: Inquiry of Supervisor
for Functional Information ............................... 118

Moderated Hierarchical Regression Results
Dependent Variable: Inquiry of Coworkers
for Functional Information ............................... 120

Moderated Hierarchical Regression Results
Dependent Variable: Observation for
Functional Information ................................. 121

Moderated Hierarchical Regression Results
Dependent Variable: Experimentation
for Functional Information ............................... 123

Moderated Hierarchical Regression Results
Dependent Variable: Inquiry of Supervisor
for Evaluative Information ............................... 124

Moderated Hierarchical Regression Results
Dependent Variable: Inquiry of Coworkers
for Evaluative Information ............................... 126

Moderated Hierarchical Regression Results
Dependent Variable: Experimentation
for Evaluative Information ............................... 127

Moderated Hierarchical Regression Results

Supervisor and Coworker Information-Sharing,
and Leader-Member Exchange (LMX) ...................... 130

Functional Information-Seeking and Mastery ..................... 136

Table 30 Evaluative Information-Seeking and Mastery ..................... 139

Table 31 Functional Information-Seeking and Innovation ................... 142
Table 32 Evaluative Information-Seeking and Innovation ................... 144
Table 33 Mastery, Innovation, and Workgroup Influence .................... 147

LIST OF FIGURES

Page
Figure 1. Conceptual Model of Newcomer Information-Seeking ................ 27
Figure 2. Strategy x Content x Tenure Interaction ......................... 96
Figure 3. Inquiry of Supervisor (Functional) x Tenure Interaction .............. 138

THEORIES OF NEWCOMER SOCIALIZATION

Socialization is the process by which a newcomer acquires the knowledge,
skills, behaviors and attitudes necessary to assume an effective role in an organization
(Feldman, 1981; Van Maanen & Schein, 1979). The newcomer must "learn the
ropes" of his/her role to become a contributing organizational member (Schein, 1968).
Individuals who enter a new organization face an uncertain situation, and they often
experience "reality shock" (Hughes, 1958) or "surprise" (Louis, 1980) when the
situation does not meet their expectations. During socialization, newcomers learn to
make sense (Louis, 1980) of the work environment in order to interpret how they fit
into the organization (Van Maanen and Schein, 1979). Theories of organizational
socialization have typically been concerned with stages of socialization (e.g., Feldman,
1981; Graen, 1976; Van Maanen, 1976), or organizational tactics for socialization
(Van Maanen and Schein, 1979). Only more recently has the role of the individual as
an active participant in the socialization process been detailed (e.g., Jones, 1983;
Louis, 1980; Miller & Jablin, 1991; Morrison, 1991, 1993; Ostroff & Kozlowski,
1992; Reichers, 1987).
W

Some authors have described the socialization process as a series of stages
through which a newcomer must pass to become assimilated into the organization

(e.g., Feldman, 1981; Graen, 1976; Porter, Lawler, & Hackrnan, 1975;

2
Van Maanen, 1976). Stage models of socialization have typically involved three

stages of newcomer adjustment (Fisher, 1986).

The first stage, the "anticipatory" stage, consists of all newcomer learning prior
to joining the organization (Feldman, 1981). This stage would include occupational
choice, organizational selection, and acceptance of the job offer (Fisher, 1986).
Newcomer expectations about what the new role will be like also develop at this time
(Wanous, 1980).

The "encounter" stage (Porter et aL, 1975), occurs when the newcomer first
begins his/her job in the organization. During this stage, newcomers learn and master
the tasks required of their job. They learn what their role will be in this new setting.
They also develop interpersonal relationships with their supervisor and peers (Fisher,
1986). Finally, they discover if their expectations match the organizational reality
(Wanous, 1980).

The final stage, or "change and acquisition" period (Porter et a1., 1975), refers
to the processes by which the newcomer becomes accepted and assimilated into the
organization (Fisher, 1986). During this stage, the newcomer masters the skills
required for the job, adjusts to workgroup values and norms, and becomes successful
in the performance of his/her role (Feldman, 1981). The newcomer becomes an
"insider," learning how things really work in the organization (Fisher, 1986).

Research has been mixed in its support of stage theories of socialization
(Fisher, 1986). Feldman (1976) has provided some support for stage theories. He
found that job satisfaction and group influence was greater for individuals who had

completed all three stages of socialization than for those who had not completed all

3
three or any stages at all. Graen, Orris, and Johnson (1973) have also presented some

evidence for the validity of stage models. By studying the activities that clerical
employees engaged in while first on the job, they learned that "assimilation behaviors"
were high during the first 16 weeks. However, these behaviors steadily decreased
over time, while attempts at resolving conflicts increased over time. This transition
into different types of newcomer behaviors may be indicative of separate stages of
socialization.

On the other hand, research has indicated that it is difficult to distinguish
discrete stages of socialization (Fisher, 1986). Even in the Graen et a1. (1973) study,
the behaviors of the clerical employees changed steadily over time, rather than
beginning or ending suddenly. In his research on police officer socialization, Van
Maanen (1975) also reported smooth changes in motivation and commitment, rather
than abrupt changes that would indicate transitions into different stages. In sum, stage
models are useful in describing some of the processes involved in newcomer
socialization, but their specific sequencing of events may not generalize to all
situations.

Socialization Tactics

Socialization has also been described in the literature as the tactics used by the
organization to shape employees and bring them "up to speed." For example, Wheeler
(1966) examined the influence of individual v. collective socialization, and serial v.
disjunctive socialization. Evan (1963) studied the effects of being socialized alone or

with a group of peers. Van Maanen (1978) and Van Maanen and Schein (1979) have

4
described six dimensions on which organizational socialization tactics may differ. The
six dimensions are:

(1) collective v. individual socialization processes;

(2) formal v. informal socialization processes;

(3) sequential v. random socialization processes;

(4) fixed v. variable socialization processes;

(5) serial v. disjunctive socialization processes; and

(6) investiture v. divestiture socialization processes.

According to Van Maanen and Schein (1979), an individual may respond to a
given role with custodianship, content innovation, or role innovation. Custodianship
occurs when the newcomer accepts the status quo and conforms to past ways of
accomplishing tasks. A newcomer who is content innovative accepts the mission of
the role, but changes the knowledge or strategies used to accomplish role
requirements. Finally, a role innovator rejects the norms of a given role; he/she
redefines the knowledge, strategies, and mission associated with the role.

Van Maanen and Schein (1979) contend that the manner in which an
organization combines the dimensions of socialization tactics will result in
custodianship, content innovation, or role innovation on the part of the newcomer. For
example, they posit that socialization processes that are individual, informal, random,
disjunctive, and involve investiture will lead to role innovation by the newcomer.
Empirical evidence has supported some of these assertions (Allen & Meyer, 1990;
Jones, 1986).

However, there are several problems with this conceptualization of socialization

and its applicability to the experience of newcomers. As Van Maanen and Schein

(1979) explain, some dimensions pertain to the socialization that occurs throughout an

5

individual’s career with an organization. For example, sequential vs. random
socialization processes refer to the degree to which the organization specifies discrete
steps in attaining an organizational role. However, these processes are more
applicable to an individual’s progression through hierarchical boundaries into higher-
level positions (Van Maanen and Schein, 1979). This dimension of socialization may
not have a large influence on a newcomers’ acceptance of their first role in the
beginning months of the job.

Furthermore, processes that are applicable to newcomers, such as formal vs.
informal processes, and individual vs. collective, may lose their influence once the
person is on the job. For instance, even if a person receives formal, collective training
before entering the job role, further socialization of the newcomer takes place when
that individual is placed in a workgroup and begins to learn his/her actual place in the
organization. The newcomer’s interactions with coworkers and supervisors will also
impact his/her adaptation to the organization.

Most importantly, Van Maanen and Schein’s (1979) theory did not account for
the effects of individual differences, nor did it acknowledge that there may be an
interaction between organizational influences and individual differences. This theory
views newcomers as passive recipients of whatever actions the organization takes to
shape them into the types of employees desired. The possibility that individual
differences in terms of personality or disposition might also impact the role orientation

taken on the job is not addressed.

An Interactionist Pemtive

Newcomers to an organization bring with them, among other things, individual
differences in personality (Fisher, 1986), expectations (Louis, 1980; Wanous, 1980),
need for control and need for feedback (Nicholson, 1984), and past socialization
experiences (Katz, 1980). While organizational strategies will impact newcomer
adjustment, the dispositions, motivations, and attitudes of the individuals to be
socialized must also be considered. Jones (1983) has noted, however, that research
rarely considers the role of individual differences in the socialization process.

Louis (1980) describes socialization as a process through which individuals
make sense of their new situations. Individuals have certain expectations about their
jobs, their coworkers, and the organization before entering the organization as a new
employee (Wanous, 1980). Surprise occurs when there is a discrepancy between a
newcomer’s expectations and their actual experiences in the new setting (Louis, 1980).
Surprise is a normal part of organizational entry, and sense-making is the process
through which individuals cope with this surprise and adapt to the new setting. The
ability of the newcomer to explain these surprises and to predict their future
occurrence will determine the successful socialization of that new member.

Newcomers have special sense-making needs, such as assistance in interpreting
events in the new setting, as well as help in understanding and accepting context-
specific interpretation schemes. Louis (1980) notes that insiders are a potentially rich
source of assistance for newcomers. This implies an interactionist view on the
socialization process, in which interactions between newcomers and insiders determine

the successful socialization of new members.

7

Jones (1983) also took an interactionist view of the socialization process, in
which both the organization and the individual influence the outcomes of socialization.
Specifically, Jones (1983) hypothesized that an individual’s self-efficacy and
attributional processes moderate that individual’s adjustment to the work role. For
example, Jones (1983) posited that individuals high in self-efficacy may tend to take a
proactive stance towards role performance, whereas those low in self-efficacy may be
more likely to accept role definitions offered by others.

In an empirical study, Jones (1986) integrated his view of the influence of self-
efficacy with Van Maanen and Schein’s (1979) theory of organizational strategies for
socialization. Jones (1986) hypothesized that a newcomer’s level of self-efficacy
would moderate the relationship between organizational socialization tactics and
individual role orientation. Jones (1986) found that when an organization uses
institutionalized socialization strategies, custodial role orientations are more likely
among those low in self-efficacy than those high in self-efficacy.

Additional research provides evidence of the proactive efforts of newcomers to
deal with the novel environment. Feldman and Brett (1983) examined individual
coping strategies employed to deal with the uncertainty of the new job situation. The
coping strategies most favored by new hires are: getting others to provide task help;
seeking out social support; and seeking out information. Further, the use of these
strategies are stable over time (Feldman & Brett, 1983).

A number of socialization theories describe newcomer attempts to be
innovative on the job. Several individual and situational factors that influence

newcomer innovation and personal development during socialization have been

8
described (Brett, 1984; Nicholson, 1984; West & Farr, 1989). These theories indicate

that individual factors such as desire for control, desire for feedback (Nicholson,
1984), self-efficacy, anxiety (Brett, 1984), growth need strength, and task specific
skills (West & Farr, 1989) will determine the extent to which a newcomer shows
innovation and creativity on the job.

However, there are also job characteristics that will constrain or provide
opportunities for innovation, including discretion, job or task novelty, unpredictability,
and challenge (Brett, 1984; Nicholson, 1984; West & Farr, 1989). Finally, situational
factors such as group support, cohesiveness, leadership support, feedback, and norms
for innovation have also been suggested to influence newcomer innovation (West &
Farr, 1989). Thus, these theories consider the interaction between the individual
entering the new situation and the characteristics of the situation in determining
socialization outcomes.

Reichers (1987) advocates the interactionist perspective by asserting that the
rate of newcomer socialization is dependent on the frequency of symbolic interactions
between the newcomer and insiders. Symbolic interactionism is a process of verbal
and social interaction through which shared meanings develop (Reichers, 1987).
Symbolic interactions occur more frequently when both newcomers and insiders are
proactive in seeking out interactions. Proaction is "any behavior that involves actively
seeking out interaction opportunities" (Reichers, 1987, p. 281). The more frequently
these symbolic interactions occur, the quicker the newcomer adjusts to the new setting.

Newcomer and insider proaction in seeking out interactions are influenced by

individual and situational variables. Individual difference variables that influence

9

proaction include field dependence, tolerance of ambiguity, and need for affiliation.
Situational factors that may affect proaction are task interdependence, orientation
programs, on-the-job training, early performance evaluations, and formal mentoring
relationships (Reichers, 1987).

In conclusion, recent socialization theories have taken into account the
interaction between the newcomer and the situation that results in adjustment to the
job. Implicit in all these theories is that newcomers are learning about the job, role,
and group when they enter the work setting. There is also some indication of the
active strategies used by newcomers to learn about the job (e.g., seeking information,
Feldman & Brett, 1983; interacting with organizational members, Reichers, 1987).
Therefore, these recent theories are more appropriate than traditional stage theories or
theories of socialization tactics in understanding the dynamic socialization process.
These theories characterize how newcomers are active in facilitating their socialization
to the new work setting. The following sections describe several theories and lines of
research that indicate in more detail _l_10_w newcomers are proactive in learning their

tasks, their roles, and their place in the workgroup.

A CONCEPTUAL MODEL OF NEWCOMER LEARNING AND ADAPTATION

In an integration of the training and socialization literatures, Feldman (1989)
explained the similarities and differences between the foci of study for the two
research domains. Training research has been concerned with evaluating the
effectiveness of formal procedures to teach individuals how to perform work tasks.
Socialization research has focused more on how newcomers learn their roles, and how
they come to understand organizational values and culture (Feldman, 1989).

Socialization research also indicates that newcomers learn quite a bit about
their tasks through their interactions with workgroup members on the job. However,
one fault of socialization research is that, in stressing that socialization is more than
just learning one’s job, it has increasingly ignored this aspect of adaptation by
focusing on the learning of attitudes and values (Feldman, 1989). Research needs to
examine more fully the processes by which newcomers learn their jobs, and how this
learning interacts with the learning of roles and organizational norms (Feldman, 1989).
Socialization Outcomes

A number of different outcomes of the socialization process have been
identified, including satisfaction, motivation, and commitment (Van Maanen, 197 5),
mutual influence, job involvement, and internal work motivation (Feldman, 1976), and
carrying out role assignments dependably, remaining with the organization, and
innovating and cooperating spontaneously (Feldman, 1981). However, the outcomes

described in conceptual discussions often differ from the outcomes measured in

10

l 1
research (Fisher, 1982). While theoretical papers highlight the learning aspect of

socialization, empirical papers often measure attitudinal outcomes of socialization
(i.e., commitment and satisfaction).

Three socialization outcomes that are related to the process of learning the job
are mastery (e.g., Falcione & Wilson, 1988; Feldman, 1981), innovation (e.g.,
Nicholson, 1984; West & Farr, 1989), and workgroup influence. Feldman (1981)

describes two process variables, resolution of role demands, and task maste_ry, that

 

reflect mastery of the job. Mastery includes learning the tasks of the new job,
determining task priorities and time allocation, dealing with conflicts, and performing
the job well (Feldman, 1981).

Furthermore, Fisher (1982) performed a qualitative study to determine what
newcomers perceived to be the outcomes of socialization. Characteristics of adjusted
newcomers included: skill proficiency; independence in action; knowledge of routines;
role clarity; and having a "system" (Fisher, 1982). These characteristics can also be
considered indicators that a newcomer has "mastered" his/her job. Thus, m can
be defined as the extent to which the newcomer learns his/her tasks, successfully and
independently performs his/her tasks and duties, and develops methods for getting
his/her work done.

A second outcome of the learning process is innovation. Innovation has been
examined as an outcome of the socialization tactics used by the organization (Allen &
Meyer, 1990; Jones, 1986; Van Maanen & Schein, 1979). It is also the result of the
newcomer’s proactive attempts to change role requirements (Nicholson, 1984). As the

newcomer learns the tasks required of him/her on the job, he/she may attempt to

12

institute better methods for getting the job done. Thus, innovation can be defined as
the implementation of new and different work objectives, methods, and procedures in
a newcomer’s job (West & Farr, 1989).

A third socialization outcome is workgroup influence. This can be defined as
the extent to which the newcomer’s supervisor and coworkers accept and use the
newcomer’s advice and suggestions, and the extent to which the workgroup includes
the newcomer in decisions. It indicates that the workgroup accepts the newcomer as a
competent performer of his/her job. Feldman (1981) indicates that making creative
suggestions and being innovative are important outcomes of socialization. "Lack of
influence, and overconforrnity to organization norms, are perhaps the most frequently
cited indicators of ineffective socialization" (Feldman, 1981, p. 315).

Feldman (197 6) found that newcomers were unwilling to make suggestions for
changes in work-related activities until they had mastered their own jobs. They felt
they needed to prove their competence before giving advice or making suggestions to
other workgroup members. Thus, it is expected that both mastery and innovation will
precede and affect the amount of workgroup influence a newcomer has. When
newcomers have mastered their jobs and show greater innovation, then workgroup
members may be more likely to take newcomer advice and include the newcomer in
decisions.

The literature on newcomer socialization has been fairly complete in describing
the outcomes of newcomer learning. However, research depicting how newcomers
mactivgly seek to achieve these outcomes has been sparse. Recently, socialization

researchers have focused on the proactive information-seeking tactics used by

13
newcomers (e.g., Ashford & Cummings, 1985; Miller & Jablin, 1991; Morrison, 1991,

1993; Ostroff & Kozlowski, 1992). Research on information sources and content (e. g.,
Greller & Herold, 1975; Hanser & Muchinsky, 1978) and proactive feedback-seeking
(Ashford, 1986; Ashford & Cummings, 1985) provide a basis for examining newcomer
proaction in seeking information about the job.

Newcomer Information Seeking

The socialization perspective taken in the present study is that newcomers are
active participants in their adaptation to their new job. Newcomers take a proactive
role in their socialization by seeking out information relevant to learning about the job
(Miller & Jablin, 1991; Morrison, 1991, 1993). Ashford and Cummings (1983) argue
that individuals are not passive recipients of feedback from their environment, but
instead actively attend to and seek information relevant to personal goals. For
example, Arps (1917) found that individuals who received no knowledge of results
from the experimenter developed their own subjective methods for ascertaining how
well they were performing in an experiment. Further, researchers have indicated that
individuals actively attend to information from a variety of sources and weigh the
importance of each source in providing information (Greller & Herold, 1975).

Ashford and Cummings (1983) explain that individuals are motivated to seek
feedback because of its value as an individual resource in attaining goals. Feedback
provides two types of information to an individual: functional information, or the
behaviors that are appropriate to achieve a goal; and evaluative information, or how
well that individual is performing the behaviors (referred to as referent and appraisal

information, respectively, by Ashford & Cummings, 1983).

14

One function of feedback is to reduce uncertainty. In fact, some degree of
uncertainty is necessary for feedback to be a valuable resource for an individual
(Ashford & Cummings, 1983). Seeking information is one way that newcomers can
proactively reduce the uncertainty of the new environment. Feedback may also serve
as a cue to which goals are valued in the environment. Finally, feedback is a
necessary resource to achieving mastery over an environment. In order to understand
the environment and make self-evaluations, the individual will attend to and seek out
relevant information (Ashford & Cummings, 1983).

Entry into a new organization is one situation that is particularly uncertain for
the individual (Feldman & Brett, 1983; Louis, 1980). Newcomers may experience
"information deprivation", in that their communication environments are limited in the
amount of useful information provided to them (Jablin, 1984, p. 622). Proactive
information-seeking is a strategy by which the newcomer can reduce the uncertainty of
the work setting (Miller & Jablin, 1991). Current socialization theory and research has
begun to examine the proactive information-seeking behaviors of newcomers (Miller &
Jablin, 1991; Morrison, 1991, 1993; Ostroff & Kozlowski, 1992).

Information content. New employees are initially faced with learning how to
perform the job (Feldman, 1976; Fisher, 1986). During socialization, newcomers must
develop the skills and knowledge necessary to job performance (Morrison, 1991,
1993). Two types of information relevant for this learning are functional information
and evaluative information (Morrison, 1991, 1993).

Functional information is information about how to perform the job, including

how to complete tasks, use equipment and tools, allocate time to various job

15
assignments, and establish work priorities (Feldman, 1981; Greller & Herold, 1975;

Miller & Jablin, 1991; Morrison, 1991, 1993; Ostroff & Kozlowski, 1992).
Newcomers may proactively seek functional information to discover what is required
of them on the job and how to accomplish it (Miller & Jablin, 1991).

Evaluative information is information concerning how well or how poorly an
individual is performing tasks and meeting job requirements (Ashford & Cummings,
1983; 1985; Greller & Herold, 1975; Miller & Jablin, 1991). Seeking evaluative
information is vital to learning about the job, since it allows newcomers to identify
and correct any errors in their performance, as well as discover areas where additional
learning is needed (Ashford & Taylor, 1990).

Information-seeking strategies. In addition to seeking different types of
information, newcomers may choose different strategies for acquiring this information.
One strategy that a newcomer can use is inquiry, which is directly asking others for
information (Ashford & Cummings, 1983). Supervisors and coworkers would be the
most relevant parties for newcomers to ask for functional and evaluative information.
For example, Louis, Posner and Powell (1983) found that coworkers and supervisors
are the most helpful sources during newcomer adaptation.

Furthermore, in her study of newcomer information-seeking, Morrison (1991)
found that other sources such as indirect superiors, support staff, friends, family
members, and clients were sought infrequently for information. However, newcomers
sought technical information to a great extent from supervisors and coworkers
(Morrison, 1991). Thus, for this study, inquiry will be limited to inquiry of

supervisors and inquiry of coworkers.

16

Another strategy for seeking information involves observation of other peeple
in the work setting. This strategy is the same as Ashford and Cummings’ (1983)
strategy of monitoring, in which newcomers attend to the behavior of others in the
environment, and the reactions of others to their own behavior. In addition,
newcomers may watch others performing their jobs to learn how to perform the job
and to find out how well they are doing on the job.

A fourth strategy that is closely tied to observation is experimentation.
Through their own trial and error performance of their tasks, newcomers learn how to
function on the job. All newcomers would necessarily participate in this strategy
because it comprises the performance of their tasks. However, some newcomers may
rely more on their own initiative in learning the tasks rather than asking supervisors or
coworkers for assistance. Furthermore, newcomers may try to evaluate how well they
are doing on the job through attending to and evaluating their own performance, rather
than relying on others’ feedback and evaluations.

Based on Bandura’s (1977b) social learning theory, individuals learn through
observation and modeling of others’ behaviors. Thus, it is likely that the strategies of
observation and experimentation will occur together. The combination of observation
and experimentation has been found to be an effective learning strategy; for example,
behavior modeling has been found to be a successful training technique (e.g., Gist,
Schwoerer, & Rosen, 1989; Goldstein & Sorcher, 1974). Individuals use the strategies
of observing and experimenting in their own proactive attempts to understand the new

job. Research indicates that newcomers engage in high levels of monitoring for

17

information (Morrison, 1991), and that both observation and experimentation are
positively related to knowledge about the job (Ostroff & Kozlowski, 1992).

Strategy choice. Newcomers may choose which information-seeking strategy
to use based on various costs associated with these strategies. For example, inquiry,
observation, and experimentation require effort costs. Inquiry involves the physical
effort of finding and asking others for information, while observation and
experimentation require the cognitive and attentional effort of processing the
information observed and then modeling it (Ashford & Cummings, 1983).

A second cost related to strategy choice is the cost of "losing face" when using
inquiry as an information-seeking strategy (Ashford & Cummings, 1983; Miller &
Jablin, 1991; Morrison & Bies, 1991). Communication with others involves social
costs (Miller & Jablin, 1991). Newcomers may be reluctant to ask supervisors or
coworkers for information for fear of appearing incompetent. Therefore, impression
management concerns may limit the newcomer’s willingness to inquire of others for
information (Morrison & Bies, 1991).

Inference costs may also affect which strategy is chosen (Ashford &
Cummings, 1983). Inquiry may require less inference, since the message is directly
conveyed to the newcomer. On the other hand, observation and experimentation
require a great deal of inference to interpret the meaning of the behaviors displayed by
others. Ftu'ther, newcomers may have inadequate interpretation schemes for attaching
meaning to events in the new environment (Louis, 1980). Thus, the use of inquiry
may provide the newcomer with more accurate information about the environment

than information acquired through observation and experimentation.

18

The importance of these costs in choosing an information-seeldng strategy may
depend on the information content. For example, one of the first tasks faced by the
newcomer is learning the requirements of his/her job (Feldman, 1981; Fisher, 1986;
Ostroff & Kozlowski, 1992). Therefore, functional information would be critical when
the newcomer is first learning his/her tasks. Without functional information, the
newcomer may not be able to perform adequately (Morrison, 1991, 1993).

Newcomers may be less constrained in their choice of strategies to seek this
information, because the cost of not knowing this information would be greater than
the costs mentioned above.

On the other hand, asking for evaluative information may have high social
costs for the newcomer. The newcomer is not performing at the level of an
experienced incumbent, and they may not want to draw attention to that fact (Morrison
& Bies, 1991). However, evaluative information is still important to their mastery of
the job. Therefore, they may use observation and experimentation to a high degree to
acquire evaluative information, while limiting their use of inquiry for evaluative
information (e.g., Morrison, 1991).

Research on Information-Seeking

There is a fairly substantial body of research that sheds some light on the
information-seeking process, although some studies present conflicting results. Several
studies examined the inforrnativeness of various sources of information for job
incumbents (Greller & Herold, 1975; Hanser & Muchinsky, 197 8). Greller & Herold
(1975) found that. for functional information, supervisors were the most informative

source, followed by the self, task, coworkers, and the company. For evaluative

19

information, the most informative source was the self, followed by the task,
supervisor, coworker, and company. Hanser and Muchinsky (197 8) found similar
results, except that, for functional as well as evaluative information, the self and task
were most informative, followed by supervisor, coworkers, and the organization.

In contrast, Burke & Bolf (1986) found that coworkers were more valuable
sources of information in general than the supervisor. However, in terms of
information content, supervisors were more valuable sources of job requirements and
evaluative information, while coworkers were more valuable sources of technical
information. Finally, individuals who were younger and at lower levels in the
organization rated coworkers as more valuable learning sources (Burke & Bolf, 1986).

Several studies examined characteristics that affect information search and
choice of information source. For example, Gerstberger and Allen (1968) found that
individuals used information sources that were more accessible. Heslin, Blake, and
Rotton (1972) found that a combination of response importance and uncertainty
(insufficient, inconsistent, or unfavorable information) led to greater information
search. Furthermore, Greller and Herold (197 5) examined several self, task, and
relational characteristics that affected reliance on information sources. These results
will be detailed in subsequent sections.

The research of Ashford and her colleagues clarifies some of the antecedents of
active information search (Ashford, 1986; Ashford & Cummings, 1985). For example,
Ashford (1986) found that the greater the importance of performance as a goal, the
greater the uncertainty about performance-related issues, and the lower the job tenure,

the more the individual valued feedback on performance. The greater the perceived

20
value of feedback, the greater the frequency of both monitoring and inquiring for

performance feedback (Ashford, 1986). Furthermore, Ashford and Cummings (1985)
found that greater uncertainty and role ambiguity led to greater information seeking,
moderated by the individual’s tolerance for ambiguity (Ashford & Cummings, 1985).

Several studies have examined the information acquisition of newcomers in
organizations (Jablin, 1984; Morrison, 1991, 1993; Ostroff & Kozlowski, 1992).
Jablin (1984), analyzing communication logs kept by newcomers, found that nurses
did not initiate most of their interactions with others during their third week of
employment. However, by the ninth week of employment, newcomers were initiating
the majority of their interactions with others (J ablin, 1984).

Ostroff and Kozlowski (1992) examined four types of information acquired by
newcomers: task; role; group; and organization. They found that newcomers acquired
the most information about the task domain, followed by role, group, and
organizational domains, and this pattern remains stable over time. This indicates that
newcomers concentrate primarily on learning about their tasks and role during their
first year on the job (Ostroff & Kozlowski, 1992).

Ostroff and Kozlowski (1992) also examined newcomer reliance on various
sources of information. For task information, supervisors and coworkers provided
information to an equal degree. In terms of noninterpersonal sources, newcomers
acquired more task information from experimentation than from observing others.
Across all four content areas, observation is relied on to the greatest extent to acquire
information, followed by supervisors and coworkers. However, there is a significant

increase in experimentation over time. Finally, although the amount of information

21

acquired from observation is nearly the same as that acquired from supervisors and
coworkers, only information acquired through observation and experimentation are
related to knowledge about the task domain (Ostroff & Kozlowski, 1992).

Morrison (1991) examined newcomer efforts to acquire different types of
information, using Ashford and Cummings’ (1983) two information-seeking strategies,
inquiry and monitoring. She found that when seeking performance feedback (i.e.,
evaluative information), newcomers used monitoring more frequently than inquiry. In
fact, they monitored for performance feedback more frequently than any other type of
information. When seeking technical (i.e., functional) information, newcomers used
inquiry more fi'equently than monitoring. Furthermore, newcomers inquired of
supervisors more frequently than coworkers for technical information and performance
feedback (Morrison, 1991).

Morrison (1993) also examined the relationship between information-seeking
and task mastery. Results indicated that inquiring for performance feedback, and
asking supervisors, asking coworkers, and monitoring for technical information were
related to task mastery. However, the relationship between asking coworkers for
technical information and mastery was negative. Morrison (1993) explained that
newcomers who are more unsure about their task competence would more frequently
ask coworkers for technical information to cope with their uncertainty.

In conclusion, research has indicated that newcomers seek information through
a number of different strategies. Further, they may rely to a greater extent on certain
strategies for certain types of information. The following section describes several

hypotheses to be tested in the present study.

22

Differential Use of Information—Seeking Strategies

The present study will examine which information strategies are used by
newcomers during their first year on the job. Based on previous research, it is
expected that newcomers will rely on information-seekin g strategies to varying
degrees. The present study hypothesizes that newcomers will rely to the greatest
extent on both observation and experimentation to acquire functional information.
Newcomers are faced with a new, uncertain situation, but they may not want to appear
incompetent. Therefore, they may rely on their own observations and behaviors to
seek information while they learn how to interact with their supervisor and coworkers.

Previous research indicates that coworkers are valuable sources of functional
information (Burke & Bolf, 1986; Louis, et al., 1983). In addition, newcomers may
feel it is less costly to ask coworkers than supervisors for functional information.
Inquiry of coworkers for functional information may be the next most frequently used
strategy, followed by asking supervisors for functional information. Morrison (1991)
investigated the hypothesis that inquiry of coworkers would be used more frequently
than inquiry of supervisors to acquire technical (i.e., functional) information. Her
results, however, indicated that the newcomers in her sample inquired of supervisors
more frequently than coworkers for technical information. Monison (1991) explained
that the relationship between supervisors and newcomers was probably unique in her
sample, in that supervisors were accountants like the newcomers and were only two
years ahead of the newcomers. In the present study, which samples a variety of jobs,
it is expected that newcomers will ask coworkers for functional information more

frequently than they ask their supervisors for this information.

23

It is expected that observation and experimentation will also be the most
frequently used strategies for acquiring evaluative information. Evaluative information
may be viewed as much more costly by newcomers, who are naturally bound to make
mistakes while learning. However, it is important for newcomers to understand how
their supervisor evaluates their work; therefore, it is expected that, for evaluative
information, newcomers will inquire of supervisors more frequently than coworkers.

In addition to the expectation that information-seeking strategies will be used
with varying frequency, it is expected that the frequency of using certain strategies
may change over time. For example, newcomers may increase their use of inquiry
over time for evaluative information. The pattern of information acquisition exhibited
by newcomers may differ over the course of their first year on the job. The present
study will compare the information-seeking strategies used by a cross-sectional sample
of newcomers on the job for approximately one year or less.

It is expected that individuals on the job for a longer period of time will have
learned the basics of how to perform the job. For example, Graen et al. (197 3) found
that over sixteen weeks on the job, individuals decreased their involvement in
"assimilation-relevant" activities, including learning the amount of work needed to be
done, mastering the job situation, and getting help from others. Furthermore, research
indicates that predictors of performance change over time (J ablin, 1984). At six weeks
of employment, the best predictor of a newcomer’s performance is the extent to which
he/she is dependent on the supervisor for information. By the eighteenth week of
employment, the best predictor of an individual’s performance is the supervisor’s

supportiveness and degree of upward work influence (Jablin, 1984).

24

Therefore, it is expected that the use of inquiry of both supervisors and
coworkers will decrease with increased tenure on the job. On the other hand, it is
expected that all newcomers in the present study will still acquire functional
information through observation and experimentation to a high degree. While it is not
a specific hypothesis that can be tested, it is expected that the use of observation and
experimentation for functional information will not change with increased job tenure
(for individuals on the job for one year or less).

In addition, while acquiring functional information from others may become
less important, discovering how others evaluate their work may gain increasing
salience for individuals who have been on the job for a longer period. These
individuals have passed through the "honeymoon period" (Miller & Jablin, 1991), and
their workgroup may expect more from them. Furthermore, they may be facing
performance evaluations by their supervisors. For example, Feldman (1977) found
that new employees usually took about 6 months to feel competent on the job, and this
usually occurred because of a feedback meeting with the supervisor. Therefore,
inquiry of both supervisors and coworkers for evaluative information is expected to
increase with greater tenure on the job. Again, while not a testable hypothesis, it is
expected that the use of observation and experimentation for evaluative information

will not change with job tenure.

25

To summarize, the hypotheses to be tested are as follows:

Hymthesis 1. A three-way interaction between information content (functional,
evaluative), information-seeking strategy, and time on the job will affect
information-seeking.

Hymthesis 2. Newcomers will use experimentation more frequently than
inquiry of coworkers and inquiry of supervisors to acquire functional
information.

Hymthesis 3. Newcomers will use observation more frequently than inquiry of
coworkers and inquiry of supervisors to acquire functional information.

Hypothesis 4. Newcomers will inquire of coworkers for functional information
to a greater extent than inquiring of supervisors for functional information.

Hymthesis 5. Newcomers will use experimentation more frequently than
inquiry of supervisors and inquiry of coworkers to acquire evaluative
information.

Hypothesis 6. Newcomers will use observation more frequently than inun of
supervisors and inquiry of coworkers to acquire evaluative information.

Hypothesis 7. Newcomers will inquire of supervisors for evaluative
information to a greater extent than inquiring of coworkers for evaluative
information.

Hymthesis 8. Inquiry of supervisors for functional information will decrease
with greater time on the job.

Hypgthesis 9. Inquiry of coworkers for functional information will decrease
with greater time on the job.

Hypothesis 10. Inquiry of supervisors for evaluative information will increase
with greater time on the job.

Hymthesis 11. Inquiry of coworkers for evaluative information will increase
with greater time on the job.

The purpose of the present study is to further clarify the process of newcomer

information-seeking for functional and evaluative information. While previous

26

research has indicated general patterns of information acquisition, it has not examined
factors that may cause newcomers to differ individually in the strategies they rely on
to acquire information. There are a number of factors that may impact the
information-seeking strategies used by newcomers. Several individual, job, and
situational characteristics are posited to influence the information-seeking behaviors of
newcomers. Figure 1 presents a general heuristic of the process of newcomer
information-seeking. This conceptual heuristic will not be tested as a complete causal
model. Rather, it serves to depict the critical variables to be examined in this study.
The following sections detail the specific relationships in the model that will be
examined and tested in this study.
Individual Difference Factors

Several individual-level variables have been posited to affect the information-
seeking strategies used by new employees. For example, in their theory of adaptation
to transitions, Ashford and Taylor (1990) hypothesize that individuals who are high
self-monitors may be especially sensitive to the need for adaptation to a new
environment. Furthermore, they believe that the breadth of the individual’s past
experiences may facilitate the recognition of pertinent cues in the new setting (Ashford
& Taylor, 1990). Several other individual differences have been hypothesized to
affect inforrnation—seeking behaviors, including tolerance for ambiguity, self-esteem,
cognitive complexity, and self-efficacy (Ashford & Cummings, 1985; Ashford &
Taylor, 1990; Major & Kozlowski, 1991; Miller & Jablin, 1991).

Research has examined how some individual differences relate to information-

seeking. For example, Greller and Herold ( 197 5) found that individual involvement in

.osxomwéoszoE. 52.8262 .6 .322 3:58:00 .F 9:9”.

 

 

 

 

 

 

 

 

 

27

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

>omo_=m-=ow-.
mwozmmmuua
#59202.
zo_._.<>Ozz_
5:985 -
EonEoo -
e 85383.25 -
mozmsdz i mo_5_mm5<m<zo m9.
$89302, 2980.
3296..
GZ_¥mmw-ZO_._.<S_IO.-..Z_ IIIIIII' mmcmzoxw hmDEm§-Emm._H
_ mocmcoxm
>Iwkw<2 hmflemfiubwflmmnd ..
sz_._.<._mm nSOmvaOB
@3260.
.8535-

Gz_m<Im-ZO_._.<§mOu_Z_
mDOm EXEC;

 

 

 

28

and enjoyment of work were related to greater reliance on intrinsic sources (e.g., task
and self) for functional information. Ashford and Cummings (1985) found that
individuals who reported greater job involvement engaged in greater levels of
information seeking. They also found that tolerance for ambiguity moderated the
relationships between uncertainty, role ambiguity and level of information seeking
(Ashford & Cummings, 1985).

However, research has not examined how the individual’s past experiences
relate to their information-seeking behaviors on the new job. Self-efficacy is a
particularly relevant variable because it examines an individual’s confidence on the
new job based on prior experiences.

Self-efficacy. As depicted in Figure l, self-efficacy is posited to influence the
strategies chosen by newcomers to acquire information on the job. Bandura (1982)
describes self-efficacy as "judgments of how well one can execute courses of action
required to deal with prospective situations" (p. 122). Self-efficacy expectations are
an individual’s estimates that given behaviors will result in certain outcomes; these
expectations are based on successful performance in the past (i.e., mastery), and past
exposure to models and other vicarious experiences (Bandura, 1977a).

Mastery experiences are the most influential sources of efficacy, followed by
vicarious experiences, verbal persuasion, and information on an individual’s
physiological state (Bandura, 1982). Self-efficacy can be distinguished from self-
esteem because self-efficacy concerns task-specific expectations (Gist, 1987). For the
newcomer, self-efficacy can be defined as his/her assessment of how well he/she can

perform on the job based on past experiences and past performances.

29

Research indicates that self-efficacy is related to subsequent performance (Gist,
1987). For example, self-efficacy has been found to be positively related to the level
of goals chosen and task performance in research on goal setting (Locke, Frederick,
Lee, & Bobko, 1984). Furthermore, higher self-efficacy prior to training has led to
greater mastery of computer skills (Gist, Schwoerer & Rosen, 1989).

Jones (1983) has posited that self-efficacy plays an important role in the
socialization process. He hypothesizes that individuals high on self-efficacy would
tend to take a more proactive stance towards their role performance due to beliefs in
their personal competence. On the other hand, individuals low on self-efficacy are
unsure about their ability to handle the new job. They may be more likely to accept
prior definitions of how to get the work accomplished (Jones, 1983). In a study
examining the effects of organizational socialization tactics on role orientation, Jones
(1986) found that self-efficacy moderated this relationship. Individuals low on self-
efficacy were more likely to take a custodial orientation towards their job, while high
self-efficacy individuals were more likely to be innovative (Jones, 1986).

Self-efficacy may influence the proactive efforts of newcomers to learn about
their jobs. For example, self-efficacy may affect the newcomer’s choice of
information-seeking strategies. The strategies chosen, however, would depend on the
content of the information sought. When seeking functional information, it is
hypothesized that individuals high on self-efficacy may rely to a greater extent on
observation and experimentation. Thus, they may rely on their own past experiences
to develop methods to accomplish their tasks. They may be more independent when

learning their tasks, and seek less guidance from others (Major & Kozlowski, 1990).

30
On the other hand, individuals low on self-efficacy may be less self-sufficient.

Because they are less confident in their ability to master the job, they may be more
likely to ask others for help in learning their tasks. Therefore, they would be more
likely to ask supervisors and coworkers for functional information. In fact, Major and
Kozlowski (1990) found newcomers low on self-efficacy to be more active in seeking
out work-related interactions with their coworkers and supervisor.

When seeking evaluative information, the opposite relationship may hold.
Specifically, individuals low on self-efficacy would be more likely to seek evaluative
information through the less public strategy of observation and experimentation. They
may avoid asking for evaluative information to protect their self-concept (Larson,
1989; Northcraft & Ashford, 1990), or they may be concerned with impression
management and how others view their competence (Morrison & Bies, 1991).

The motivation to avoid damage to their public image may be particularly
salient for low self-efficacy individuals. However, acquiring evaluative information is
still important for detecting and correcting any errors in performance; therefore,
individuals low on self-efficacy may rely on observation of others and self-evaluations
to acquire this information. Although the information acquired through observation
and experimentation may be biased (Northcraft & Ashford, 1990), it is expected that
for low self-efficacy newcomers, the costs of appearing incompetent are greater.

On the other hand, individuals high on self-efficacy are expected to use inquiry
to a greater extent to acquire evaluative information. They may be more likely to
view information from others, whether positive or negative, as a way to improve their

knowledge and skills. They will tend to have less doubt about their competence, so

31
they will not be concerned that attention will be drawn to their inadequacies. In fact,

their impression management concerns may be opposite to those of low self-efficacy
individuals. By asking for evaluative information, they can direct attention to their
success at mastering the job. In this manner, they can improve their public image
through inquiring of others for evaluative information (Morrison & Bies, 1991).
Furthermore, individuals who seek evaluative information from others may be viewed
as conscientious and willing to improve their skills (Ashford & Cummings, 1983).
Research has given some indication that self-efficacy may influence the
willingness of individuals to ask for evaluative information. For example, when task
performance was manipulated to be high on a preliminary task, individuals chose to
perform more diagnostic tasks (i.e., tasks that provided more information about their
ability) in the second experimental session, compared to individuals with low initial
performance (Zuckerman, Brown, Fischler, Fox, Lathin, & Minasian, 1979).
Furthermore, Northcraft and Ashford (1990) examined the willingness to ask
for evaluative information more directly. In a laboratory study, they manipulated
performance expectations prior to the experiment with an "aptitude test. " They found
that individuals with low performance expectations inquired for less evaluative
information, especially when inquiry was public (N orthcraft & Ashford, 1990).
Though not labelled self-efficacy, these experiments did manipulate task-specific
performance expectations based on past experiences, which fit the definition of self-
efficacy. These experiments provide some support for the notion that self-efficacy

influences the strategies used to acquire evaluative information.

32

In the present study it is expected that self efficacy will moderate the
relationship between job characteristics and information-seeking strategies. Therefore,
the specific hypotheses concerning self-efficacy will be detailed in the next section.
Job Characteristics

From an interactionist perspective, both individual and situational
characteristics will influence the newcomer adaptation process (Schneider, 1983). For
example, the type of job the newcomer enters may influence the newcomer’s learning
strategies. It is posited here that certain job characteristics may constrain or provide
opportunities for newcomers to utilize different information-seeking strategies. Job
characteristics may be a salient feature of the environment for newcomers, because
newcomers focus early on task learning (Morrison, 1993; Ostrofi & Kozlowski, 1992).
Figure 1 details three job characteristics posited to play a role in the newcomer
adaptation process: interdependence; complexity; and discretion.

Interdmndence. In Thompson’s (1967) theory of technology, the
technological requirements of an organization will dictate the type of interdependence
found between organizational parts. There are three types of interdependence that can
be found in organizations. In order of increasing interdependence they are: 1) pooled
interdependence; 2) sequential interdependence; and 3) reciprocal interdependence.
Pooled interdependence describes the situation in which each component of an
organization contributes to the whole organization and is supported by the whole
organization. Sequential interdependence occurs when the outputs of one part of the

organization become the inputs to another part of the organization. Finally, when the

33

outputs of one part of the organization become the inputs of the other, and vice versa,
the situation is described as reciprocal interdependence (Thompson, 1967).

These three types of interdependence exhibit increasing degrees of contingency.
As the level of interdependence increases, there is a greater necessity for interaction
and communication between organizational members to accomplish tasks (Thompson,
1967). Research at the workgroup level has found a relationship between
interdependence and coordination modes (Van de Ven, Delbecq, & Koenig, 197 6).
Van de Ven et a1. (1976) found that as interdependence increased between workgroup
members, increases in the use of coordination modes such as horizontal channels of
communication and unscheduled and scheduled meetings occurred.

Interdependence functions similarly at the individual level of analysis.
Individual jobs can have differing degrees of interdependence with other jobs within a
workgroup. This may then impact the level of communication between workgroup
members. The level of interdependence between a newcomer’s job and the jobs of
other workgroup members impacts the opportunities the newcomer has to ask
coworkers for information. In fact, interdependence is an important component of
Reicher’s (1987) theory of newcomer proaction. She asserts that greater
interdependence between jobs may facilitate proaction because it necessitates
interactions between newcomers and insiders (Reichers, 1987).

Major and Kozlowski (1990) tested this hypothesis and found that newcomers
initiated greater levels of work-related interactions with their coworkers when their
jobs were more interdependent. To tie these results more specifically to utilization of

information-seeking strategies, it can be hypothesized that. newcomers in jobs that are

34

more interdependent will use inquiry as a strategy to acquire information to a greater
extent. Since interdependent jobs require interactions between individuals to
accomplish tasks, newcomers will have more opportunities to ask others for the
information they need.

On the other hand, newcomers who are in jobs that are not interdependent will
have fewer opportunities to interact with coworkers. In some cases, coworkers may
have less knowledge about the newcomer’s job when there is little interdependence
between jobs. In addition, where newcomers and coworkers are performing similar
duties yet are not interdependent with one another, newcomers may be hesitant to
bother their coworkers too frequently with questions. Therefore, newcomers in jobs
that have little interdependence may rely on observation and experimentation to a
greater extent to acquire information. It is expected that interdependence will have
similar impacts on strategies to acquire both functional and evaluative information.

In addition, it is expected that self-efficacy will moderate the relationships
between interdependence and the various information-seekin g strategies. For
individuals low on self-efficacy, it is expected that a strong positive relationship will
be found between interdependence and inquiry of supervisors and coworkers for
functional information. On the other hand, for individuals high on self-efficacy, there
should be a weaker relationship between interdependence and inquiry of supervisors
and coworkers. As found by Major and Kozlowski (1990), individuals low on self-
efficacy may be more sensitive to the situational demand of high interdependence.
Since they are not confident about their ability to succeed on the job, they are already

more likely to ask others for assistance in determining the correct ways to perform

35

their job. The opportunity for more frequent interactions with coworkers due to
greater interdependence will facilitate their inclination to rely on inquiry to learn their
jobs. On the other hand, individuals high on self-efficacy may be less sensitive to the
effect of interdependence because they are more confident that they will succeed in
learning and performing their tasks.

Furthermore, for individuals low on self-efficacy, the hypothesized negative
relationships between interdependence and observation and experimentation for
functional information should be found. Individuals low on self-efficacy should be
more likely to rely on observation and experimentation when their job is not
interdependent with coworkers and they have less opportunities to ask questions. In
contrast, for high self-efficacy individuals, there should be no relationship between
interdependence and observation and experimentation for functional information. High
self-efficacy individuals should engage in relatively higher degrees of observation and
experimentation regardless of the level of task interdependence.

Hypothesized interactions are slightly different in terms of seeking evaluative
information. For individuals low on self-efficacy, it is expected that interdependence
will have a negative relationship with inquiry of supervisors and coworkers for
evaluative information. Individuals low on self-efficacy already believe that they are
not performing well on the job; therefore, the social costs of appearing incompetent
make them less likely to inquiry for evaluative information. When their jobs are
interdependent with others, they may be less likely to inquire for evaluative
information because they will be highlighting their lack of success on the job to those

individuals who must depend on their work. For individuals. high on self-efficacy,

36
there should be the hypothesized positive relationship between interdependence and

inquiry of supervisors and coworkers for evaluative information. Individuals who are
more confident of their abilities are more likely to ask for evaluative information to
improve their performance, and greater interdependence provides more opportunities
for inquiry.

Furthermore, individuals low on self-efficacy should report little relationship
between interdependence and observation and experimentation for evaluative
information. Low self-efficacy individuals are expected to rely on observation and
experimentation to acquire evaluative information to a greater extent regardless of the
level of interdependence. For individuals high on self-efficacy, the hypothesized
negative relationship between interdependence and observation and experimentation for
evaluative information is expected.

The specific hypotheses to be tested are as follows:

Hypothesis 12. Task interdependence will be positively related to inquiry of
supervisors for functional information.

Hypothesis 12a. This relationship will be moderated by the newcomer’s
self-efficacy. For low self-efficacy individuals, a strong, positive
relationship between task interdependence and inquiry of supervisors for
functional information is expected; for high self-efficacy individuals this
relationship should be weaker.

Hypothesis 13. Task interdependence will be positively related to inquiry of
coworkers for functional information.

Hypothesis 13a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, a strong, positive relationship between task
interdependence and inquiry of coworkers for functional information is
expected; for high self-efficacy individuals this relationship should be
weaker.

37

Hypothesis l4. Task interdependence will be negatively related to observation
for functional information.

Hypothesis 14a. Newcomer self-efficacy will moderate this
relationship. For individuals with low self-efficacy, there should be a
negative relationship between task interdependence and observation for
functional information. For high self-efficacy individuals, task
interdependence should have no relationship to observation for
functional information.

Hypothesis 15. Task interdependence will be negatively related to
experimentation to acquire functional information.

Hypothesis 15a. Self-efficacy will moderate this relationship. For
individuals with low self-efficacy, there should be a negative
relationship between task interdependence and experimentation for
functional information. For high self-efficacy individuals, task
interdependence should have no relationship to experimentation for
functional information.

Hypothesis 16. Task interdependence will be positively related to inquiry of
supervisors for evaluative information.

Hymthesis 16a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, task interdependence will be negatively related
to inquiry of supervisors for evaluative information. For high self-
efficacy individuals, task interdependence will be positively related to
inquiry of supervisors for evaluative information.

Hypothesis 17. Task interdependence will be positively related to inquiry of
coworkers for evaluative information.

Hyppthesis 17a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, task interdependence will be negatively related
to inquiry of coworkers for evaluative information. For high self-
efficacy individuals, task interdependence will be positively related to
inquiry of coworkers for evaluative information.

38

Hyppthesis 18. Task interdependence will be negatively related to the use of
observation to acquire evaluative information.

Hypothesis 18a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, interdependence should have no relationship to
observation for evaluative information. For individuals high on self-
efficacy, there should be a negative relationship between
interdependence and observation for evaluative information.

Hyppthesis 19. Task interdependence will be negatively related to the use of
experimentation to acquire evaluative information.

Hymthesis 19a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, interdependence should have no relationship to
experimentation for evaluative information. For individuals high on
self-efficacy, there should be a negative relationship between
interdependence and experimentation for evaluative information.
Con_rplexig. Figure 1 indicates a second job characteristic hypothesized to
affect newcomer information seeking, complexity. There have been numerous
definitions of complexity in the literature. For example, March and Simon (1958)
define complex tasks as those tasks having uncertain alternatives and consequences of
action, inexact means-end relationships, and a number of smaller subtasks that make
up the complete task. Terborg and Miller (197 8) indicate that more complex tasks are
high on "path-goal multiplicity;" in other words, a number of strategies for
accomplishing the task exist. Wood (1986) describes three components of complexity.
Component complexity is the number of distinct acts to be performed and distinct
information cues to be processed in order to complete a task Coordinative
complexity concerns the relationships between inputs and products, and includes

timing, frequency, intensity and location requirements for task performance. Finally,

dynamic complexity concerns changes in the means-end relationships over time.

39
After reviewing the literature, Campbell (1988) describes four task

characteristics that contribute to complexity:

(a) the presence of multiple potential ways (i.e., paths) to arrive at a desired

end-state, (b) the presence of multiple desired outcomes (i.e., end-states) to be

attained, (c) the presence of conflicting interdependence among paths to
multiple outcomes, and (d) the presence of uncertain or probabilistic links

among paths and outcomes (Campbell, 1988, p. 43).

The various definitions described above all define complexity as an objective
task characteristic. This definition of complexity can be differentiated from task
difficulty, which represents a person X task interaction (Campbell, 1988). For
example, a complex task may be easy or difficult to perform based on an individual’s
past experiences. One problem with the definitions described above are their own
"complexity;" especially in the case of Wood’s (1986) and Campbell’s (1988)
definitions, it may be difficult to measure all aspects of complexity described.

On the other hand, Perrow (1967) describes two dimensions of an
organization’s technology that are part of the construct of job complexity. One
dimension is the number of exceptions, or task variety, that occur in the performance
of work. A second dimension of complexity is the analyzability of the job. When
tasks are highly analyzable, individuals can rely on established procedures to get the
job done. However, when tasks are not very analyzable, individuals must think about
how to solve problems on the job, and they may have to engage in active search
beyond available procedures (Withey, Daft, & Cooper, 1983). The two dimensions

can be combined to form a continuum from routine (low on both dimensions) to

nonroutine technologies (high on both dimensions). To be more exact, the two

40

dimensions can be combined to form four categories of technologies: routine, craft,
engineering, and nonroutine (Perrow, 1967).

Research examining the link between complexity and communication indicates
that, at the organizational level of analysis, more complex organizations exhibit greater
levels of horizontal communication between departments (Hage, Aiken, & Marrett,
1971). Furthermore, in analyzing communication within work units, it was found that
the more complex the work unit’s task, the greater the level of communication
between coworkers (Tushman, 197 9).

Daft and Macintosh (1981) examined the relationship between the two
dimensions of complexity and the information-processing within work units. They
found that the greater the number of exceptions (or task variety), the greater the
amount of information processed by the workgroup. Furthermore, information
equivocality was greater when tasks were less analyzable. A puzzling finding was that
greater amounts of information were processed when tasks were more analyzable.
They hypothesized that information may be difficult to share when activities are
intangible and difficult to understand, and it may be more difficult to develop coding
schemes to interpret information about less analyzable tasks. Furthermore, they
hypothesized that individuals in less analyzable jobs may use more equivocal sources
of information, such as engaging in face-to-face interactions to obtain information
(Daft & Macintosh, 1981).

The dimensions of complexity described at the work unit level are also relevant
and meaningful at the individual job level. The present research will examine

complexity at the individual job-level. Drawing on the research described above, it is

41
expected that job complexity will impact the information-seeking strategies utilized by

newcomers. One dimension of complexity, exceptions, has been related to the use of
information sources. Greller and Herold (1975) found that individuals who reported
greater variety in job duties relied to a greater extent on extrinsic sources (supervisor
and coworkers) for functional information.

Additional hypotheses concerning the relationships between job complexity and
information-seeking can be made. If a job is fairly routine, the newcomer may not
need to actively inquire for information (Ashford & Cummings, 1983). When there
are few exceptions, or the job is analyzable, the individual may be able to rely to a
great extent on observation and experimentation to master his/her job. On the other
hand, if a job has many exceptions or is not analyzable, the individual may not be able
to learn the requisite information needed for job mastery on his/her own. Thus,
newcomers entering more complex jobs may inquire of supervisors and coworkers to a
greater extent for information.

It is expected that job complexity will have the same impact for both functional
and evaluative information. The newcomer will engage in greater levels of inquiry for
both functional and evaluative information when exceptions are high. On the other
hand, when there are few exceptions, the newcomer may be more self-sufficient in
learning the job, relying to a greater extent on observation and experimentation.
Similarly, when the job has low analyzability, the newcomer will be more likely to
inquire for functional and evaluative information. When the job is highly analyzable,
the newcomer may rely on observation and experimentation to acquire functional and

evaluative information.

42
These relationships, however, are hypothesized to be moderated by self-

efficacy. Specifically, for individuals low on self-efficacy, the hypothesized
relationships between exceptions, analyzability, and inquiry of supervisors and
coworkers for functional information are expected. When individuals have low self-
efficacy, greater task variety and lower analyzability will lead to more frequent inquiry
of supervisors and coworkers for functional information. In contrast, for individuals
high on self-efficacy, there will be a weaker relationship between complexity and
inquiry of supervisors and coworkers for functional information. Individuals who are
confident of their abilities will be less likely to inquire for functional information
overall. Their confidence is expected to lessen the impact of complexity on inquiry.
There will be some information that they must acquire from others, but they are
expected to react to greater variety and less analyzability more proactively with greater
experimentation.

Thus, high self-efficacy individuals will report a positive relationship between
exceptions and observation and experimentation for ftmctional information. They
should also be more likely to use observation and experimentation for functional
information when the job is less analyzable. For low self-efficacy individuals, the
hypothesized negative relationships between exceptions and observation and
experimentation for functional information are expected. In addition, it is expected
that individuals with low self-efficacy will be more likely to observe and experiment
for functional information when the job is more analyzable. When it is less

analyzable, they do not have the confidence or experience to learn their tasks on their

own.

43

In terms of evaluative information, low self-efficacy individuals are expected to
inquire less of supervisors and coworkers when the job has more exceptions, and when
the job is less analyzable. When individuals are faced with more complex jobs, and
they have lower self-efficacy, they may avoid appearing incompetent to other
workgroup members by refraining from asking for evaluative information. Individuals
high on self-efficacy, however, may feel challenged by jobs with greater exceptions
and less analyzability, and may therefore be more willing to ask for evaluative
information to improve their skills.

Low self-efficacy individuals faced with complex jobs may also be less likely
to observe and experiment for evaluative information. Individuals are more likely to
experience mistakes and failures in a complex job, and low self-efficacy individuals
will be less willing to confirm their expectations through active self-evaluation. On
the other hand, high self-efficacy individuals should engage in observation and
experimentation for evaluative information more frequently when their job has more
exceptions and less analyzability. The greater challenge of a complex job will lead
them to more actively assess their skills and abilities as a means of improvement.

The following hypotheses will be tested in this study:

Hypothesis 20. Exceptions will be positively related to inquiry of supervisors
for functional information.

Hwothgis 203. Self—efficacy will moderate this relationship.
Exceptions will be positively related to inquiry of supervisors for
functional information for individuals low on self-efficacy. For
individuals high on self-efficacy, the relationship should be weaker.

44

Hymthesis 21. Exceptions will be positively related to inquiry of coworkers
for functional information.

Hyppthesis 21a. Self-efficacy will moderate this relationship.
Exceptions will be positively related to inquiry of coworkers for
functional information for individuals low on self-efficacy. For
individuals high on self-efficacy, the relationship should be weaker.

Hymthesis 22. Exceptions will be negatively related to observation to acquire
functional information.

Hyppthesis 22a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, exceptions should be negatively related to
observation to acquire functional information. For high self—efficacy
individuals, exceptions should be positively related to observation for
functional information.

Hymthesis 23. Exceptions will be negatively related to the use of
experimentation to acquire functional information.

Hypothesis 23a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, exceptions should be negatively related to
experimentation to acquire functional information. For high self-
efficacy individuals, exceptions should be positively related to
experimentation for functional information.

Hymthesis 24. Exceptions will be positively related to inquiry of supervisors
for evaluative information.

Hymthesis 243. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, exceptions will be negatively related to inquiry
of supervisors for evaluative information. For high self—efficacy
individuals, exceptions will be positively related to inquiry of
supervisors for evaluative information.

Hymthesis 25. Exceptions will be positively related to inquiry of coworkers
for evaluative information.

Hypothesis 25a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, exceptions will be negatively related to inquiry
of coworkers for evaluative information. For high self-efficacy
individuals, exceptions will be positively related to inquiry of coworkers
for evaluative information.

45

Hymthesis 26. Exceptions will be negatively related to the use of observation
to acquire evaluative information.

Hypothesis 26a. Self-efficacy will moderate this relationship.
Exceptions will be negatively related to observation for evaluative
information for low self-efficacy individuals. Exceptions will be
positively related to observation for evaluative information for high self-
efficacy individuals.

Hypothesis 27. Exceptions will be negatively related to the use of
experimentation to acquire evaluative information.

Hypothesis 27a. Self-efficacy will moderate this relationship.
Exceptions will be negatively related to experimentation for evaluative
information for low self-efficacy individuals. Exceptions will be
positively related to experimentation for evaluative information for high
self-efficacy individuals.

Hymthesis 28. Analyzability will be negatively related to inquiry of
supervisors for functional information.

Hypothesis 28a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, analyzability will be negatively related to
inquiry of supervisors for functional information. For high self-efficacy
individuals, the relationship will be weaker.

Hypothesis 29. Analyzability will be negatively related to inquiry of coworkers
for functional information.

Hyppthesis 29a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, analyzability will be negatively related to
inquiry of coworkers for functional information. For high self—efficacy
individuals, the relationship will be weaker.

Hymthesis 30. Analyzability will be positively related to the use of
observation to acquire functional information.

Hymthesis 30a. Self-efficacy will moderate this relationship.
Analyzability will be positively related to observation for functional
information for low self-efficacy individuals. The relationship will be
negative for high self-efficacy individuals.

45

Hypothesis 31. Analyzability will be positively related to the use of
experimentation to acquire functional information.

Hypothesis 31a. Self-efficacy will moderate this relationship.
Analyzability will be positively related to experimentation for functional
information for low self-efficacy individuals. The relationship will be
negative for high self-efficacy individuals.

Hyppthesis 32. Analyzability will be negatively related to inquiry of
supervisors for evaluative information.

Hypothesis 32a. Self-efficacy will moderate this relationship. For
individuals low on self-efficacy, analyzability will be positively related
to inquiry of supervisors for evaluative information. For high self-
efficacy individuals, the relationship will be negative.

Hymthesis 33. Analyzability will be negatively related to inquiry of coworkers
for evaluative information.

Hypothesis 33a. Self-efficacy will moderate this relationship. For
individuals low on self-efficacy, analyzability will be positively related
to inquiry of coworkers for evaluative information. For high self-
efficacy individuals, the relationship will be negative.

Hymthesis 34. Analyzability will be positively related to the use of
observation to acquire evaluative information.

Hypothesis 34a. Self-efficacy will moderate this relationship.
Analyzability will be positively related to observation for evaluative
information for low self-efficacy individuals. For high self-efficacy
individuals, the relationship will be negative.

Hymthesis 35. Analyzability will be positively related to the use of
experimentation to acquire evaluative information.

Hypothesis 35a. Self-efficacy will moderate this relationship.
Analyzability will be positively related to experimentation for evaluative
information for low self-efficacy individuals. For high self-efficacy
individuals, the relationship will be negative.

47

Discretion. A third job characteristic that may impact newcomer information-
seeking is discretion (Brett, 1984; Nicholson, 1984). Discretion (i.e., autonomy) can
be defined as an individual’s freedom to determine the methods for performing the
job, to alter goals, to schedule work, to act independently of his/her supervisor, and to
decide whom to deal with in order to get work accomplished (Brett, 1984; Hackrnan &
Oldham, 1980; Nicholson, 1984; West & Farr, 1989). Labelled "autonomy" in the job
redesign literature, it is believed to be one of five job characteristics that can improve
internal work motivation (Hackrnan & Oldham, 1980). Autonomy is hypothesized to
relate to the psychological state of experienced responsibility for work outcomes:

When the job provides substantial autonomy to the persons performing it, work

outcomes will be viewed by those individuals as depending substantially on

their Pl“. efforts, initiatives, and decisions, rather than on, say, the adequacy of
instructions from the boss or on a manual of job procedures (Hackrnan &

Oldharn, 1980, p. 79).

A meta-analysis by Spector (1986) found that autonomy on the job was related
to greater levels of satisfaction on the job, lower levels of role ambiguity, higher
performance, lower intent to quit and turnover, and higher motivation. In a subsequent
meta-analysis of job characteristics, Fried and Ferris (1987) found that autonomy was
related to growth satisfaction, and experienced meaningfulness and responsibility.

In the socialization literature, discretion is described as a factor that allows
newcomers the opportunity to innovate when developing their roles within the
organization (Brett, 1984; Nicholson, 1984; West & Farr, 1989). Nicholson (1984)

asserts that jobs high in discretion ". . . make it impossible simply to conform to job

specifications, role descriptions, or the practices of previous incumbents: newcomers

48

lack adequate data on which to base any such conformity; therefore, some degree of
role development is inevitable" (Nicholson, 1984, p. 178). Several studies have
supported the relationship between job discretion and innovation (West, 1987; West,
Nicholson, & Rees, 1987).

West (1987) performed a longitudinal study of the levels of innovation
displayed by a sample of managers and professionals. He found discretion to predict
innovation at two points in time, as well as predict changes in innovation over time.
In a study of managers and professionals who had moved into newly-created jobs,
West, et a1. (1987) found that these individuals reported significantly greater levels of
discretion than individuals moving into existing jobs. Although the relationship
between discretion and innovation was not examined in this study, it can be
hypothesized that an individual moving into a newly created role is implicitly a role
innovator (West, et al., 1987).

With regard to newcomer adaptation, it can be posited that the level of
discretion in the job and the level of innovation displayed by the newcomer are linked
by the strategies chosen by the newcomer to acquire information. Specifically, it can
be hypothesized that individuals who enter jobs with greater levels of discretion will
be more likely to use observation and experimentation to learn information about the
job (i.e., functional information).

When a job is high in discretion, the newcomer may be forced to develop
his/her own methods to get tasks accomplished. Furthermore, his/her supervisor and
coworkers may not provide much help in determining the best ways to get work done.

For example, West, et al. (1.987) found that individuals entering newly-created jobs

49

were less likely to report that their bosses, subordinates, and a colleague were helpful
in letting them know about the requirements of their new job. Furthermore, Greller
and Herold (1975) found that individuals who reported higher levels of job discretion
relied to a greater extent on intrinsic sources (task and self) for functional information.

On the other hand, when a job is low in discretion, the newcomer is allowed
little latitude in developing his/her own work methods. Therefore, the newcomer may
rely to a greater extent on his/her supervisor and coworkers to learn about job
requirements. Other workgroup members are more likely to shape the behavior of
newcomers in low discretion jobs to ensure that they behave in specified ways.
Therefore, it would be expected that newcomers in low discretion jobs would be more
likely to inquire of supervisors and coworkers for functional information.

In terms of evaluative information, it is expected that job discretion will have a
positive relationship with all strategies for acquiring evaluative information. When
individuals are in high discretion jobs, there are no standard procedures for getting the
work done. Newcomers placed in high discretion jobs are faced with ambiguity in
terms of how well they are performing their tasks and duties. Therefore, newcomers
in high discretion jobs may need to be more active in terms of evaluating their own
performance, as well as in discovering how supervisors and coworkers evaluate their
performance. In contrast, newcomers in low discretion jobs may have clear
instructions on how to perform job duties. Therefore, they may have less of a need
for evaluative information.

The relationships between discretion and information-seeking strategies will be

moderated by self-efficacy. For high self-efficacy individuals, the hypothesized

50

negative relationships between discretion and inquiry of supervisors and coworkers for
functional information should be found. Individuals who are more confident when
entering a new job are better able to handle the self-sufficiency required by a job with
high discretion. On the other hand, low self—efficacy individuals are expected to
inquire of supervisors and coworkers for functional information to a greater degree
when faced with more job discretion. Low self-efficacy individuals are not able to
handle the challenge of developing their own work methods, and thus are more likely
to ask others for help in determining how to best accomplish their work.

Similarly, low self-efficacy individuals will be less likely to observe and
experiment for functional information when placed in jobs with greater discretion.
They are less likely to have the skills and experience to learn their job with no
assistance. On the other hand, high self-efficacy individuals will be more likely to
observe and experiment for functional information when faced with job discretion.

With regard to evaluative information, low self-efficacy individuals will be less
likely to inquiry of supervisors and coworkers when their job is high in discretion.
The social costs of appearing incompetent will prevent them from asking for this
feedback. In contrast, individuals high on self-efficacy will be more likely to inquiry
of supervisors and coworkers for evaluative information when their job has more
discretion. When faced with the challenge of developing their own work methods, it
becomes more important to learn whether their methods are correct. Individuals high
on self-efficacy will be less threatened by the possibility of negative feedback, they

Will be more likely to view it as diagnostic.

51
For high self-efficacy individuals, there should also be a positive relationship

between job discretion and observation and experimentation for evaluative information.
When the job has greater autonomy, it is necessary for individuals to determine correct
and incorrect ways of performing their tasks. For low self-efficacy individuals, this
relationship should be weaker. Their expectation that they are not performing well
may make them less likely to evaluate their own performance and thus add to their
uncertainty.

The hypotheses to be tested are described below:

Hymthesis 36. Discretion will be negatively related to inquiry of supervisors
for functional information.

Hypothesis 36a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, discretion will be positively related to inquiry
of supervisors for functional information. For high self-efficacy
individuals, the relationship will be negative.

Hymthesis 37. Discretion will be negatively related to inquiry of coworkers
for functional information.

Hypothesis 37a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, discretion will be positively related to inquiry
of coworkers for functional information. For high self-efficacy
individuals, the relationship will be negative.

Hymthesis 38. Discretion will be positively related to the use of observation
to acquire functional information.

Hypothesis 38a. Self—efficacy will moderate this relationship.
Discretion will be negatively related to observation for functional
information for low self-efficacy individuals. For high self-efficacy
individuals, the relationship will be positive.

52

Hyppthesis 39. Discretion will be positively related to the use of
experimentation to acquire functional information.

Hypothesis 39a. Self-efficacy will moderate this relationship.
Discretion will be negatively related to experimentation for functional
information for low self-efficacy individuals. For high self-efficacy
individuals, the relationship will be positive.

Hymthesis 40. Discretion will be positively related to inquiry of supervisors
for evaluative information.

Hypothesis 40a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, the relationship between discretion and inquiry
of supervisors for evaluative information will be negative. For high
self-efficacy individuals, the relationship will be positive.

Hymthesis 41. Discretion will be positively related to inquiry of coworkers for
evaluative information.

Hypothesis 41a. Self-efficacy will moderate this relationship. For low
self-efficacy individuals, the relationship between discretion and inquiry
of coworkers for evaluative information will be negative. For high self-
efficacy individuals, the relationship will be positive.

Hymthesis 42. Discretion will be positively related to the use of observation
to acquire evaluative information.

Hymthesis 42a. Self-efficacy will moderate this relationship. For high
self-efficacy individuals, discretion will be positively related to
observation for evaluative information. For low self-efficacy
individuals, the relationship will be weaker.

Hyppthesis 43. Discretion will be positively related to the use of
experimentation to acquire evaluative information.

Hymthesis 43a. Self-efficacy will moderate this relationship. For high
self-efficacy individuals, discretion will be positively related to
experimentation for evaluative information. For low self-efficacy
individuals, the relationship will be weaker.

53
Workgoup Relations

Feldman (1989) has indicated the need for socialization research to examine
". . . how the learning of a job interacts with the learning of a role and of group
norms" (p. 405). Although these learning processes have been separated theoretically,
they all occur within the same environment and will thus interact to some degree. The
processes of job and role learning may be particularly linked due to the interactive
natures of each of the processes. While interacting with workgroup members to learn
the basics of job performance, newcomers must also learn how they fit into the
workgroup and what these other organizational members expect of them. Further,
newcomers must develop relationships with both their supervisor and coworkers in
order to become effective organizational members.

Falcione and Wilson (1988) emphasize the transactional and reciprocal nature
of the socialization process. Newcomers attend to environmental cues that they obtain
through interacting with other organizational members. These cues help them to
interpret the new environment. Both ambient (indirect and unconscious) and
discretionary (direct and intentional) cues play a role in the intersubjective process of
socialization (Falcione & Wilson, 1988).

Their theory is similar to Reicher’s (1987) perspective in which the interactions
between newcomers and insiders lead to the development of shared meanings. From
both these perspectives, it is apparent that the communication between newcomers and

their supervisors and coworkers is vital to the socialization process.

54

Whom coworker information-sharing. As depicted in Figure 1, the
willingness of supervisors and coworkers to share functional and evaluative
information will have direct impacts on newcomer information-seeking. Often,
feedback is infrequently provided in organizations (Ilgen, Fisher, & Taylor, 1979).
Ashford and Taylor (1990) and Jablin (1984) note that newcomers are often provided
with limited amounts of useful information. Thus, it may be expected that newcomers
must cope with this lack of information by asking for the information they need, or
discovering the information through observation and experimentation.

On the other hand, the willingness of supervisors and coworkers to provide
functional and evaluative information may be positively related to how active the
newcomer is in acquiring this information. This would suggest that, in workgroups
where supervisors and coworkers are proactive in providing this information, the
newcomer may perceive a norm for sharing information. In fact, research indicates
that, when individuals perceive a strong workgroup norm for sharing information (i.e.,
colleagues enthusiastically share information without being asked), individuals are
more likely to use interpersonal sources when seeking technical information (Dewhirst,
1971). Newcomers, who are actively seeking cues for how to behave in their new
environment (e.g., Louis, 1980), may pick up on these norms and behave accordingly.
In fact, Major (1990) found that the proaction of supervisors and coworkers in
providing work-related information was positively related to the proaction of
newcomers in actively seeking out this information. Therefore, the present study will
test the prediction that the frequency with which supervisors and coworkers share

functional and evaluative information will be positively related to newcomer attempts

55

to acquire functional and evaluative information through all information-seeking

strategies.

Hymthesis 44. Sugrvisor sharing and coworker sharing of functional
information will be positively related to the extent to which the newcomer

seeks functional information through inquiry of supervisors.

Hypothesis 45. Suxrvisor sharing and coworker sharing of functional
information will be positively related to the extent to which the newcomer

seeks functional information through inquiry of coworkers.

Hymthesis 46. Sumrvisor sharing and coworker sharing of functional
information will be positively related to the extent to which the newcomer
seeks functional information through observation.

Hymthesis 47. Sugrvisor sharing and coworker sharing of functional
information will be positively related to the extent to which the newcomer

seeks functional information through experimentation.

 

Hypothesis 48. Supgrvisor sharing and coworker sharing of evaluative
information will be positively related to the extent to which the newcomer
seeks evaluative information through inquiry of supervisors.

Hyppthesis 49. Supe_rvisor sharing and coworker sharing of evaluative
information will be positively related to the extent to which the newcomer

seeks evaluative information through inquiry of coworkers.

Hyppthesis 50. Sumisor sharing and coworker sharing of evaluative
information will be positively related to the extent to which the newcomer

seeks evaluative information through observation.

Hymthesis 51. Sugrvisor sharing and coworker sharing of evaluative
information will be positively related to the extent to which the newcomer

seeks evaluative information through experimentation.

Leader-member exchange relationships. During socialization, the newcomer
learns how to interact and communicate with his/her supervisor. It is particularly
important for the newcomer to learn how to deal with his/her superior, since the

newcomer is dependent on the supervisor for job assignments and instruction, rewards,

56

and for linking that newcomer to the rest of the organization (Falcione & Wilson,
1988; Katz, 1985; Jablin, 1985).

Because the supervisor is in a position of authority, he/she is able to clarify any
ambiguities experienced by newcomers in terms of task assignments, responsibilities,
or expectations (Falcione & Wilson, 1988). "Frequent and open communication with
the supervisor can provide the newcomer with more opportunities to clarify concerns,
ask questions, check perceptions, and become better acquainted with the organizational
culture" (Falcione & Wilson, 1988, p. 157). Communication not only clarifies
uncertainties about the job, but may also facilitate the development of a personal
relationship between the newcomer and supervisor (Jablin, 1982).

Dansereau, Graen, and Haga (1975) suggest that different types of exchange
relationships may develop between supervisors and subordinates. First, a supervisor
may develop an exchange relationship with a new employee based solely on the
formal authority he/she possesses as a supervisor. In this case, the newcomer agrees
to fulfill his/her role requirements as well as submit to his/her supervisor’s authority,
in exchange for formal compensation in terms of salary and benefits.

On the other hand, a relationship may develop between the newcomer and
supervisor based on social and interpersonal exchanges. The supervisor would provide
various benefits to the newcomer, such as "job latitude, influence in decision making,
Open and honest communications, support of the member’s actions, and confidence in
and consideration for the mem " (Dansereau, et al., 1975, p. 50). In return, the

subordinate would devote greater time and energy to ensuring the success of the work

57

unit, and would assume more job responsibility and become more committed to the
organization (Dansereau et al., 1975).

The development of a leader-member exchange (LMX) relationship occurs
through the processes of role taking, role making, and role routinization (Graen &
Scandura, 1987). During role taking, the supervisor gives the newcomer various job
assignments to discover the newcomer’s abilities and motivations. Role making
consists of negotiations between the supervisor and newcomer in terms of how they
will behave in various situations. During role making, the two parties exchange
resources valued by each member. Finally, during role routinization, the patterns of
behaviors between the supervisor and subordinate become more permanent and
interlocked (Graen & Scandura, 1987). The development of a higher quality exchange
relationship would be characterized by the supervisor providing greater latitude, _
information, support, and influence to the subordinate. The subordinate would exhibit
dependability, commitment, and loyalty (Dansereau et al., 1975).

Research indicates that supervisors develop different quality exchange
relationships with newcomers (Graen, Orris, & Johnson, 1973). Furthermore,
Dansereau et a1. (1975) found that supervisors developed different quality relationships
with members of the same work unit. The quality of these relationships was apparent
within the newcomer’s first month on the job. Members of the supervisor’s "in-group"
reported getting more attention than reported by "out-group" members, although "out-
group" members reported a greater preference for attention. Individuals with greater
negotiating latitude had less difficulty in dealing with the supervisor, and reported

their supervisor was more responsive to their job needs (Dansereau et al., 1975).

58

Graen and Ginsburgh (1977) found that member role orientations and leader
acceptance of members had an interacting influence on member resignations. Graen,
Cashman, Ginsburg, and Schiemann (1977) reported that supervisors who had higher
quality exchange relationships with their superiors provided greater support to their
subordinates, and these subordinates showed greater involvement in administrative
decision-making activities. They commented that the ". . . quality of upper dyads can
be viewed as a major component of the organizational climate of members" (Graen et
al., 1977). In an examination of the relationships between climate and LMX,
Kozlowski and Doherty (1989) found that members’ negotiating latitude was related to
their climate perceptions, and that in- group members showed greater consensus on
their climate perceptions and greater similarity to their supervisor’s climate perceptions
than out- group members.

Finally, Graen, Novak and Sommerkamp (1982) trained supervisors to
exchange their job expectations with those of their subordinates. Individuals whose
supervisors received this training showed greater productivity, greater motivating
potential of their jobs, greater role orientation, higher job satisfaction, greater work
overload, and lower future job ambiguity. Thus, research indicates that the
development of high-quality exchange relationships between newcomers and
supervisors may lead to a number of positive outcomes for both parties.

The role-making processes between supervisors and newcomers may be related
to the information-seeking tactics utilized by new employees to learn their jobs. Since
these two processes may overlap somewhat during newcomer adaptation, role making

and job learning will probably reciprocally influence one another (see Figure 1).

59
Several hypothesized relationships between the two processes can be specified.

Interactions between newcomers and supervisors will initially consist of task-based
information to a large degree. The extent to which newcomers ask supervisors for
functional information may influence the development of the leader-member exchange
relationship. When newcomers proactively seek functional information from their
supervisors, they are increasing the amount of communication between them.
Frequent communication is essential to the process of adaptation (Jablin, 1985), and is
also related to higher quality LMX relationships (Graen & Scandura, 1987).

In addition, newcomers who are more proactive in seeking information may be
perceived more favorably by their supervisors. For example, supervisors may perceive
newcomers who seek functional information from them as actively interested in
improving their skills and abilities. Newcomers are seen as needing support and
assistance (Feldman & Brett, 1983), and those newcomers who are more proactive in
seeking out this assistance may be perceived more positively by supervisors.

Furthermore, newcomers who actively seek functional information from their
supervisors may feel that their supervisors are more supportive of them. Kirmeyer and
Lin (1987) found that the rate of subordinate-initiated work-focused communication to
superiors was related to subordinate perceptions of social support. Thus, it may be
that newcomers who more frequently ask their supervisors for functional information
create more opportunities for the development of higher quality LMX relationships.

The process of developing LMX relationships may also influence the
information-seeking tactics utilized by newcomers. Newcomers may initially be

hesitant to ask for evaluative information while first learning the job (Morrison, 1991).

60

However, if a high-quality, supportive relationship develops between the supervisor
and subordinate, the individual may feel more comfortable asking for evaluative
information later in the adaptation process (Ashford & Taylor, 1990). For example,
Greller and Herold (197 5) found that individuals who reported that their supervisors
were friendly and approachable were more likely to rely on external sources
(supervisor and coworkers) for evaluative information. Thus, the processes of learning
the job and developing a relationship with the supervisor are hypothesized to influence
one another over time.

The present study is cross-sectional; therefore, the manner in which the these
processes interact over time cannot be examined. Although the present study
examines information-seeking at one point in time, several hypothesized relationships

between LMX and newcomer information-seeking can be specified:

Hypothesis 52. LMX quality will be positively related to inquiry of
supervisors for functional information.

Hymthesis 53. LMX quality will be positively related to inquiry of
supervisors for evaluative information.

Team-member exchange relationships. In addition to developing a relationship
with their supervisor, newcomers must learn to interact with their peers. Newcomers
must learn what their coworkers expect of them, as well as what role they will play in
the workgroup. "New members seek to interact with and observe members of relevant
groups or reference groups in order to gain acceptance into the group as well as

competence in carrying out their roles" (Louis, 1990, p. 94). Since coworkers are at

61
the same organizational level, they may be especially helpful in providing information
and clarifying uncertainties. In fact, coworkers have been reported to be the most
available and most helpful socialization agent for newcomers (Louis et al., 1983).

Coworkers can assist newcomers in interpreting surprises in the new situation
by providing them with relevant information (Louis, 1980). Communication between
newcomers and coworkers can thus facilitate newcomer adaptation. J ablin and
Sussman (1983) describe the functions of individually-oriented group communication.
This communication functions to:

(a) provide members with feedback about their self-concepts, (b) gratify needs

for affiliation, (0) share and test perspectives about social reality, ((1) reduce

organization-related uncertainty and concomitant individual feelings of anxiety,
insecurity, and powerlessness, or (e) accomplish employee-related (versus
organization-related) tasks and resolve individual or group-related problems

(Jablin & Sussman, 1983, p. 13).

Coworkers may also provide newcomers with social support that alleviates the
stress of the new work environment. For example, Evan (1963) found that newcomer
interaction with peer group members was negatively related to turnover of new
employees. Sheridan (1985) reported that turnover among new employees was
negatively related to group cohesion. Furthermore, Fisher (1985) found that coworker
emotional support was related to several socialization outcomes, including job
satisfaction, self-rated performance, professional commitment, and intention to leave.
However, as in the development of supervisor-subordinate relationships, these

supportive relationships between coworkers develop over time. Newcomers must not

only negotiate a role with their supervisors, but also with their peers.

62

Seers (1989) developed the construct of team-member exchange to describe the
development of exchange relationships among workgroup members. Team-member
exchange (I'MX) concerns the amount and pattern of reciprocation developed between
peers. The construct can be defined as ". . . the member’s perception of his or her
willingness to assist other members, to share ideas and feedback and in turn, how
readily information, help, and recognition are received from other members" (Seers,
1989, p. 119). TMX quality is related to the construct of group cohesion. However, it
differs in that cohesion is a perception of the group as a whole, while TMX is an
individual’s perception of his/her role within the group.

Team-member exchange (TMX) contrasts with leader-member exchange in that
it is not dyadic; rather, it involves the relationship that develops between an individual
and his/her workgroup as a whole. However, it is similar to LMX in that the
development of a team-member exchange relationship involves processes of social
exchange. Although research has examined the influence of the supervisor on the role
making process, new employees may spend a lot of time interacting with peer group
members when learning the job and role (Seers, 1989). Thus, coworkers may also
influence the development of the newcomer’s role in the organization. Furthermore,
Seers (1989) found TMX to be positively related to job satisfaction and performance.

Figure 1 indicates that the processes of acquiring information on the job and
developing an exchange relationship with coworkers will influence one another. Since
TMX quality develops through newcomer interactions with coworkers, it can be
posited that newcomer proaction in initiating communication will be positively related

to the quality of the TMX relationship. In a review of the group cohesion literature,

63
Lott and Lott (1965) reported that greater levels of interaction lead to greater

interpersonal attraction among group members. Although TMX is a distinct construct
from cohesiOn, a similar relationship between communication and TMX may be
hypothesized.

Initial communication between a newcomer and coworkers may focus on task
procedures, task priorities, and problems faced by the newcomer on the job (Louis,
1990). Newcomers may interact with coworkers to clarify ambiguities about their task
and role requirements (Falcione & Wilson, 1988; Louis, 1980). The extent to which
the newcomer inquires of coworkers for functional information may influence the
quality of the TMX relationship that develops. Since the newcomer is increasing the
amount of interaction with other workgroup members, he/she is increasing the
opportunities for exchange relationships to develop.

Team-member exchange quality may also impact the level of group
communications. For example, a similar construct, group cohesion, has been found to
be positively related to the level of communication in groups (Lott & Lott, 1961). By
definition, TMX allows for information and feedback to be more readily shared among
group members. Thus, the development of higher quality TMX relationships may
impact later efforts by newcomers to acquire information.

When newcomers develop high-quality relationships with their coworkers, they
may feel that it is less costly to ask their coworkers for evaluative information. High-
quality TMX relationships provide an atmosphere of open communication and
feedback for workgroup members. Some support for this relationship has been found

by Greller and Herold (1975). Individuals who reported that coworkers kept to

64

themselves relied more on internal sources (task and self) rather than external sources
(supervisor and coworkers) for evaluative information. This would also indicate the
opposite relationship to be true: when coworkers interact and assist one another, a
newcomer may more readily seek evaluative information from them.

As with the team-member exchange process, the present study is unable to test
the reciprocal influences of TMX and newcomer information-seeking processes.
However, several hypothesized relationships can still be specified:

Hymthesis 54. TMX quality will be positively related to inquiry of coworkers

for functional information.

Hypothesis 55. TMX quality will be positively related to inquiry of coworkers
for evaluative information.

To conclude, it is posited that a number of individual, job, and situational
characteristics impact the types of strategies chosen by newcomers to acquire
functional and evaluative information. Furthermore, Figure 1 indicates that newcomer
information-seeking is related to two socialization outcomes, mastery and innovation.
Outcomes of Newcomer Information Seeking

Previous socialization research has found relationships between information-
seeking and mastery (Morrison, 1993) and information acquisition and task knowledge
(Ostroff & Kozlowski, 1992). Furthermore, it has been stated that information is a
resource to achieve mastery over a situation (Ashford & Cummings, 1983). Figure 1
depicts a relationship between newcomer information-seeking and mastery. It is

hypothesized that greater levels of information-seeking by newcomers will lead to

65

greater mastery of the job. By seeking out relevant information, newcomers will
reduce the uncertainty of the new job.

Specifically, individuals who seek functional information to a greater degree
across all strategies will exhibit greater mastery of the job. However, the relationship
between inquiry and mastery may change over time. For example, once newcomers
have passed through a "honeymoon period" (Miller & Jablin, 1991), their supervisor
and coworkers may expect them to have learned the basics of the job. If these
newcomers, who have been on the job for a number of months, are frequently asking
these supervisors and coworkers for functional information, it may be an indication
that they have not mastered their jobs well. Therefore, time on the job may moderate
the relationship between inquiry and mastery.

In terms of evaluative information, it is expected that seeking this information
to a greater degree across all strategies will lead to greater mastery. By actively
acquiring this information, newcomers are able to adjust any errors in their
performance of the job (Ashford & Taylor, 1990).

As shown in Figure 1, newcomer information-seeking may also relate to the
amount of innovation displayed by the newcomer on the job. Individuals who are
innovative do not simply conform to past ways of performing the job; rather, they rely
on their own initiative to find better ways to get their work accomplished (e.g., Jones,
1983). This suggests that newcomers who are more self-sufficient in acquiring
functional information may be more innovative on the job. Specifically, it is expected
that newcomers who rely on observation and experimentation to acquire functional

information will display greater levels of innovation on the job. On the other hand,

66

inquiry of supervisors and inquiry of coworkers for functional information should have
little or no unique impact on newcomer innovation.

Newcomers who are more active in acquiring evaluative information in general
(across all strategies) may also be more innovative on the job. By asking for
evaluative information and evaluating their own performance, newcomers are able to
discover whether their attempts to perform the job are adequate, and they can then
improve any problems in their performance (e.g., Ashford & Taylor, 1990).
Newcomers who tend to be more active in acquiring evaluative information are likely
to find better ways to accomplish their jobs.

Figure 1 indicates that mastery and innovation mediate the relationship between
information-seeking and workgroup influence. However, this mediator relationship
will not be tested in the present study. Figure 1 serves as a conceptual heuristic to
display the theoretical model developed. The present study will be testing the links
displayed in the model in separate analyses. For example, two analyses will examine
the links between information—seeking and the two outcomes, mastery and innovation.
In addition, a separate analysis will test the relationships between mastery, innovation,
and workgroup influence.

Both mastery and innovation are hypothesized to affect the amount of influence
a newcomer has on his/her workgroup. Individuals who show that they have mastered
their jobs should find their workgroup more accepting of their ideas and suggestions.
In addition, those who have found better ways to accomplish their tasks should have

more influence on their workgroup; however, this relationship may be moderated by

mastery.

67

In summary, the hypotheses concerning information-seeking strategies and

outcomes are as follows:

Hymthesis 56. Individuals who seek functional information to a greater extent
(across all strategies) will exhibit greater mastery of the job.

Hypothesis 57. Time on the job will moderate the relationship between inquiry
of supervisors for functional information and mastery. With little time on the
job, inquiry of supervisors for functional information should be positively
related to mastery. With greater time on the job, inquiry of supervisors for
functional information should be negatively related to mastery.

Hypothesis 58. Time on the job will moderate the relationship between inquiry
of coworkers for functional information and mastery. With little time on the
job, inquiry of coworkers for functional information should be positively
related to mastery. With greater time on the job, inquiry of coworkers for
functional information should be negatively related to mastery.

Hypothesis 59. Individuals who seek evaluative information to a greater extent
(across all strategies) will exhibit greater mastery of the job.

Hymthesis 60. The use of observation and experimentation above and beyond
the use of inquiry of supervisors and coworkers to acquire functional
information should be positively related to innovation; the use of inquiry of
supervisors and coworkers beyond observation and experimentation to acquire
functional information should not have a unique impact on innovation.

Hypothesis 61. Newcomers who seek evaluative information to a greater
extent (across all strategies) will exhibit greater levels of innovation on the job.

Hypothesis 62. Mastery will be positively related to workgroup influence.

Hypothesis 63. Mastery will moderate the relationship between innovation and
workgroup influence. Innovation will be positively related to workgroup
influence for individuals high on task mastery; innovation should have no
relationship to workgroup influence for individuals low on task mastery.

METHOD
Smle and Procedure

Previous socialization research has often focused on individuals in one
profession or one organization at a time (e.g., Feldman, 1989). Therefore, the present
study sought to examine the socialization of newcomers in a variety of jobs and
organizations. Newcomers from seven organizations participated in the present study.
The seven organizations that participated in the study were as follows: a university;

a food manufacturer; a chemical manufacturer; a hospital supply/services company; a
general industrial manufacturing company; a power generation systems construction
company; and a telecommunications company. Through contacts in each of the
organizations, the researcher acquired the names of new hires and mailed them
questionnaires. If the organization had someone to handle the distribution of surveys.
the researcher sent the contact person the surveys to distribute to new hires.

Surveys were distributed to 1,382 newcomers who had been on the job for
approximately one year or less. A total of 449 newcomers returned completed
surveys, for a response rate of 32.5%. In addition to completing the survey, the
newcomers were asked to provide the names and office addresses of their supervisor
and up to three coworkers who would be sent surveys that contained similar questions
about the newcomer’s adaptation to the job.

Surveys were sent to 355 supervisors of the newcomers who had completed the

survey. A total of 214 supervisors returned the questionnaire, for a 60.3% response

68

69

rate. Newcomers provided names and office addresses for 849 coworkers who were
also sent surveys to complete. Surveys were returned by 490 coworkers, for a
57.7% response rate.

Survey data was collected from multiple sources to deal with the possible
method bias difficulties that have been cited in previous socialization research. A
critique of the literature on newcomer socialization is that it tends to collect self-report
data only, including measures of socialization outcomes. The present study sought to
obtain outcome measures from independent sources to deal with common method
variance. Specifically, it was decided that coworkers would be used as sources for the
three outcome measures of mastery, innovation, and workgroup influence. Coworkers
were chosen to provide the outcome measures for two reasons. First, it can be argued
that coworkers are in a good position to observe newcomers during their adaptation to
their job. For example, Louis et a1. (1983) found that coworkers were the most
available source for help during socialization. Daily interactions with peers was
available to nearly 9 out of 10 new employees (Louis et al., 1983). In addition,
coworkers are often present, and are able to frequently monitor and reinforce new
employee behavior with their power to provide informal rewards (Fisher, 1986).
Second, the use of coworkers as the independent source for the outcome measure
allowed for a newcomer-coworker matched sample size of N = 271 to test the study’s
hypotheses. Because of the numerous interactions hypothesized in this study,

statistical power was a great concern.

70

After deleting surveys with missing data, the final newcomer-coworker matched
sample to be used in this study was 253. The representation of newcomers from each
participating organization was as follows: university (N = 63); general industrial
manufacturer (N = 47); food manufacturer (N = 45); chemical manufacturer (N = 42);
hospital supply/services company (N = 24); telecommunications company (N = 18);
and power generation systems construction company (N = 14).

In addition to the variables collected as part of this study, individuals also
provided information on background characteristics. The new employees participating
in this study were 47% female, with an average age of 28 (sd = 7 years). Average
time on the job was 27.19 weeks (sd = 18.85). Average number of employees in the
workgroup was 12. In addition the average number of similar jobs in the past was
1.43 (sd = 1.35). Finally, the sample reported that both their supervisor and coworkers
were physically accessible and easy to get to. On a five-point scale from low to high
accessibility, supervisor accessibility was rated with a mean of 4.17 (sd = 1.02) and
coworker accessibility was rated with a mean of 4.43 (sd = .88). A variety of jobs
were sampled in this study as follows: clerical (11.1%); skilled/manual labor (6.3%);
computer programming (7.5%); engineering (30.4%); management (16.2%); sales
(3.2%); finance/accounting (3.2%); research/science (7.5%); and other professional
(14.6%).

The sample used for this study included a majority of professional employees
(N = 209), as well as a minority subsample of nonprofessional employees (clerical and
skilled/manual labor, N = 44). The subsamples could differ in terms of variances on

the variables examined, as well as in terms of their mean values. Therefore, a

71

MANOVA and Box’s M test were run to examine the possibility of mean differences
and the lack of homogeneity of variance on the variables of interest. To examine
whether professionals and nonprofessionals differed on variables used in the present
analyses, a one-way MANOVA was run with job status (professional, nonprofessional)
as the grouping factor. The dependent variables included all variables to be used in
the present analysis, along with background questions also collected on the survey.
These background characteristics included: age; gender; time on the job; number of
similar jobs in the past; number of coworkers in workgroup; physical accessibility of
supervisor; and physical accessibility of coworkers.

Box’s M test was run to examine whether there was homogeneity of variance
for the variance-covariance matrices across the between-subject factor of job status.
The Box’s M value was significant (x2 (465 (if) = 600.58, p < .001), rejecting the
hypothesis of homogeneity of variance. In addition, the MANOVA resulted in a
significant effect for job status (F(30, 204) = 6.17, p < .001), indicating that
professionals and nonprofessionals did differ on some of the variables of interest.

To examine which variables contributed most to differentiating the two groups,
a discriminant analysis was run. Discriminant structure coefficients, the pooled
within-groups correlations between the discriminating variables and the canonical
discriminant function, are listed in Table l. The list only includes those variables with
pooled within-groups correlations of .20 or higher. The results of the discriminant
analysis indicate that professionals and nonprofessionals differ mainly due to job
characteristics. Nonprofessionals tend to have jobs with less variety (exceptions) and

that are more analyzable. This is understandable in that nonprofessional jobs in this

72

sample were clerical and skilled manual labor jobs; these jobs are more likely to be
repetitious and analyzable. In addition, nonprofessionals tend to be older in age, to
have greater mastery of their jobs, and have greater self-efficacy. N onprofessional
jobs are less interdependent with coworker jobs, nonprofessionals have had more
similar jobs in the past, they have been on the job for a longer period of time, they are
more likely to be female, and they are less likely to ask coworkers for functional
information.

Because professionals and nonprofessionals do tend to differ on some of the
variables of interest in this study, job status will be used as a covariate in analyses,

where appropriate, to control for these differences.

Table 1

Discriminant Analysis Results for Professional/Nonprofessional Job Status

 

Pooled Within-Groups Correlations

 

Function 1
Exceptions .59
Age .38
Analyzability .37
Mastery .27
Self-Efficacy .26
Interdependence -.26
Similar Jobs in Past .26
Time on the Job .20
Gender .20
Inquiry of Coworker -.20

(Functional Info)

 

73

Measures

The measures to be used include new scales developed for this study, as well
as a number of scales previously used in the industrial/organizational psychology
literature. The newly developed information-seeking, mastery, and workgroup
influence measures were pretested on a sample of 313 working undergraduates prior to
this study. Scales were modified based on factor analysis results. Measures that were
taken from the literature were not pretested.

Job status. Respondents were coded as being in either nonprofessional (coded
0) or professional (coded 1) jobs. Because the MAN OVA and discriminant analysis
indicated that these two groups differed on some of the variables of interest in this
study, the job status variable was used as a covariate in most analyses.

Time on the job. Respondents answered the following open-ended question:
"How long have you held your current position (to the nearest week)?" Examination
of the distribution of this variable for the present sample indicated a non-normal
distribution. The distribution was positively skewed, and contained several peaks.
Therefore, this measure was trichotomized into three tenure categories as follows:

1 = 2-13 weeks; 2 = 14 to 36 weeks; 3 = 37 weeks or more.

All scales used in this study were subjected to factor analyses using the entire
sample of 449 newcomers who returned surveys. For measures collected from more
than one source, factor analyses were run only on the newcomer responses. For all
measures, the principal factors method was employed with a varimax rotation and
Kaiser criterion. In addition, where multiple factors emerged, a scree test was used to

determine the number of factors. The factor analysis results for each measure are

74
discussed below. In addition, reliability analyses were conducted for all measures

using the complete sample of 449 newcomers who returned surveys.

Newcomer information-seeking items were developed to tap four information-
seeking strategies (inquire of supervisor, inquire of coworkers, observe, and
experiment) and two content areas (functional and evaluative information).

Ingm’ of sum’sor - functional information. This seven-item scale measured
the extent to which the individual asked his/her supervisor how to perform tasks, use
equipment, and handle problems on the job. Respondents answered these items on a
five-point scale ranging from (1) very infrequently to (5) very frequently. New
employees provided responses to this measure. Items can be found in Appendix A.

Ingm‘ of sumisor - evaluative information. This seven-item scale measured
the extent to which the individual asked the supervisor how well he/she was
performing on the job and meeting job responsibilities. Respondents answered these
items on a five-point scale ranging from (1) very infrequently to (5) very frequently.
New employees responded to this measure. Items can be found in Appendix B.

Ingm' of coworkers - functional information. This seven-item scale measured
the extent to which the individual asked his/her coworkers for information on how to
perform the tasks and duties of the job. Respondents answered these items on a
five-point scale ranging from (1) very infrequently to (5) very frequently. New
employees responded to this measure. Scale items can be found in Appendix C.

Ingm’ of coworkers - evaluative information. This seven-item scale measured
the extent to which the individual asked his/her coworkers how well he/she was

performing on the job and meeting job requirements. Respondents answered these

75
items on a five-point scale ranging from (1) very infrequently to (5) very frequently.

New employees answered these items. Scale items are listed in Appendix D.

Observation - functional information. This six-item scale measured the extent
to which the individual observed the behavior of others to determine how to perform
tasks, how to use equipment and materials, and how to handle difficulties on the job.
New employees answered these items on a five-point scale ranging from (1) very
infi'equently to (5) very frequently. Scale items are listed in Appendix E.

Observation - evaluative infonnation. This six-item scale measured the extent
to which the individual observed the behavior of others to determine how well he/she
was performing on the job, and how well others evaluated his/her work. New
employees answered these items on a five-point scale ranging from (1) very
infrequently to (5) very frequently. Scale items are listed in Appendix F.

Em’ ntation - functional information. This six-item scale measured the
extent to which the individual experimented on his/her own to discover how to
perform tasks, use equipment and materials, and handle problems on the job. New
employees answered these items on a five-point scale ranging from (1) very
infrequently to (5) very frequently. Scale items are listed in Appendix G.

Exm’ ntation - evaluative information. This six-item scale measured the
extent to which the individual evaluated his/her own performance on the job. New
employees answered these items on a five-point scale ranging from (1) very
infrequently to (5) very frequently. The items can be found in Appendix H.

The 52 items making up these eight information-seeking scales were subjected

to a common factor analysis and a varimax rotation. The factor analysis results

76

suggested an eight-factor solution, although only seven interpretable factors were
rotated. The scree plot revealed that six- and seven-factor solutions were also
reasonable. Therefore, two additional factor analyses specifying six- and seven-factor
solutions were run. The seven-factor solution produced the same factors as those
found in the original factor analysis. The six-factor solution was less consistent with
the theoretical distinction between the information-seeking strategies. Therefore, based
on theoretical and content grounds, the seven-factor solution was chosen for use in the
present analyses. The factor loadings for the seven factors are presented in Table 2.
A11 factors accounted for at least 2.3 percent of the total variance.

To a large extent, the predicted factor structure was maintained. For example,
all items concerned with inquiry of coworkers for evaluative information fell on the
second factor. Factor 3 consisted of all items that measured inquiry of supervisors for
evaluative information. Factor 4 included all items concerned with inquiry of
supervisors for functional information. Factor 5 included all items tapping
experimentation for functional information. Finally, Factor 6 included all items that
measured experimentation to acquire evaluative information.

However, several factors were not as expected. For example, Factor 1
contained all observation items, both for functional and evaluative information, as well
as two items concerned with inquiry of coworkers for functional information. In
addition, two other items tapping inquiry of coworkers for functional information
loaded on Factor 2 along with the asking coworkers for evaluative information items.
Finally, the remaining three inquiry of coworkers for functional information formed

Factor 7.

Principal Axis Factor Matrix for Information-Seeking Items

77

Table 2

 

 

Factor
1 2 3 4 5 6 7
Observe Evall‘ J2 .16 .06 -.05 .00 .07 .06
Observe Eva14' .11 .23 .03 .05 .01 .08 -.06
Observe Eval3‘ 15 .14 .01 .05 .01 .12 .02
Observe FuncS ,_7_3_ .22 .08 .11 .01 .08 .14
Observe Func3 fl .17 .07 .14 -.03 .05 .24
Observe Func6 12 .19 .05 .03 .08 .04 .14
Observe ch2 fi .25 .15 .08 -.04 .07 .10
Observe Funcl fi .11 .02 .13 -.01 -.04 .27
Observe Func4 ._6; .16 .03 .09 .09 .02 .30
Observe Eval6‘ ,Q .15 .06 -.03 .11 .20 -.04
Observe Eva12‘ £6 .11 .07 -.O7 .12 .21 -.03
Observe EvalS‘ ,42 .13 .25 -.01 .15 .13 -.06
Cowk Ask ch2‘ ,fi .15 .06 .20 .03 -.06 .44
Cowk Ask Funcl‘ & .23 .02 .17 .01 -.13 .36
Cowk Ask Eva15 .26 ;8_2 .26 .08 -.01 .11 .13
Cowk Ask Eval7 .28 £2 .26 .05 .00 .04 .13
Cowk Ask Eval6 .28 ._7_8_ .31 .05 -.O3 .03 .09
Cowk Ask Eva13 .28 .7_7 .26 .05 -.02 .06 .09
Cowk Ask Eval2 .30 fl .24 .06 .02 .04 .08
Cowk Ask Eval4 .30 ,7_6 .31 .09 -.03 .07 .08
Cowk Ask Func3' .38 ._5_3 .20 .17 -.01 .03 .21
Cowk Ask ch6‘ .42 ._5§ .15 .22 -.01 -.03 .21
Cowk Ask Evall‘ .26 fig .17 .18 .00 .04 .10
Supv Ask Eva17 .09 .26 ,8] .19 -.02 .06 .02
Supv Ask Eval6 .07 .25 J .22 -.02 .08 .05
Supv Ask Eval4 .07 .20 g .22 -.02 .09 .05
Supv Ask Eva13 .09 .26 ._8_Q .23 -.01 .06 .03
Supv Ask Eva15 .09 .27 16 .24 -.02 .13 -.O7
Supv Ask Evall .14 .19 .7_5 .20 -.02 .09 .06
Supv Ask Eval2 .13 .27 5 .30 -.02 .07 .02

78
Table 2 (cont’d.)

 

Factor

 

1 2 3 4 5

Supv Ask Func3 .08 .11 .17 g 03 -.01 .08
Supv Ask Func2 .03 .ll .15 .84 - 05 .00 .03
Supv Ask Funcl .10 .04 .09 11 -.O3 -.07 .05
Supv Ask Func7 .01 .14 .18 14 -.03 -.00 .20
Supv Ask Func6 .07 .03 .23 .6_7 - O4 .02 .08
Supv Ask Func5 .04 .05 .32 :69 04 .
Supv Ask Func4‘ .09 .11 .33 ._5_1 - 05 .04 .11
Experiment Func2 .09 -.08 .02 .01 5 .
Experiment FuncS .02 -.Ol -.02 -.05 .8 .20 .06
Experiment Func6 .07 -.02 -.O2 -.06 ,8_1_ .24 .02
Experiment Func3 .03 .00 .02 -.03 1‘1 .22 -.O6
Experiment Funcl .06 -.O4 -.05 -.02 .76 .08 -.06
Experiment Func4 .04 .06 -.04 .01 . l .12 .06

Experiment Eva15 .16 .07 .08 .01 .17 _§§
Experiment Eval2 .06 .01 .06 .03 .18 11
Experiment Eva13 .10 .03 .00 .06 .18 18
Experiment Eval6 .17 .07 .10 -.03 .18 fi -.05
Experiment Eval4 .12 .09 .14 -.06 .14 _7_1
Experiment Evall .02 -.02 .06 -.03 .17 p;

Cowk Ask Func4 .44 .29 .02 .20 .01 -.07 ,6;
Cowk Ask FuncS .36 .33 .05 .26 -.02 --.06 £4;
Cowk Ask Func7 .39 .36 .07 .23 -.00 -.04 6

EIGEN VALUE 15.03 6.25 5.20 3.37 2.51 1.88 1.19

 

‘ These items were deleted from the final information-seeking scales used in this
study.

79

It appeared from these factor analyses that newcomers do not distinguish the
content of the information they acquire through observation. However, the present
study predicted distinct hypotheses depending on the content of the information
observed. Based on the conceptual ideas developed in this paper, it was decided that
the present study would only focus on the use of observation to acquire functional
information. Upon reflection, it was decided that observation for functional
information was more clear and understandable as a construct than observation for
evaluative information (i.e., it seems more difficult for individuals to obtain
evaluations of their work by monitoring the behavior of others). Therefore, it was
decided that all items concerning observation for evaluative information would be
dropped from Factor 1.

In addition, although four items measuring inquiry of coworkers for functional
information loaded on different factors, there was a distinct three-item factor that did
measure this construct. Therefore, based on analysis of item content, theoretical
considerations, and a factor loading criteria of .54 or greater, several items were
dropped from three of the factors to maintain the conceptual distinction between the
constructs. Specifically, three items were dropped from Factor 1, three items from
Factor 2, and one item from Factor 4 based on these criteria. In addition, as
mentioned above, the five items concerned with observation for evaluative information
that loaded above .54 were also dropped from Factor 1 to maintain the conceptual
clarity of the constructs. The items that were dropped from the factors are marked in

Table 2. This resulted in seven clearly interpretable factors.

80

The seven measures, their number of items, and reliability coefficient (for the
complete sample of 449 newcomers returning surveys) are as follows: (1) inquiry of
supervisors for functional information (six items, or = .90); (2) inquiry of supervisors
for evaluative information (seven items, a = .95); (3) inquiry of coworkers for
functional information (three items, a = .91); (4) inquiry of coworkers for evaluative
information (six items, or = .96); (5) observation for functional information
(six items, or = .90); (6) experimentation for functional information (six items,

0'. = .92); and (7) experimentation for evaluative information (six items, a = .90).

Self-efficacy. Self-efficacy was measured with a six-item scale adapted from
Major (1990). This scale measured newcomer confidence in performance on the job
based on past experiences. Four items comprising the Major (1990) scale, as well as
two new items consistent with this scale, made up the present self-efficacy measure.
The rating scale for this measure ranged from (1) strongly disagree to (5) strongly
agree. The items for this scale can be found in Appendix I.

The six self-efficacy items were subjected to a factor analysis using the
principal factors method. One six-item factor was extracted, which accounted for
67.4 percent of the variance. The factor loadings for the items can be found in
Table 3. In addition, the internal consistency reliability estimate for the self-efficacy

scale was .90.

8 1
Table 3

Principal Axis Factor Matrix for Self-Efficacy Items

 

Factor
1
Self-Efficacy6 .87
Self-EfficacyS .84
Self-Efficacy4 .80
Self-Efficacy2 .73
Self-Efficacy3 .73
Self-Efficacyl .70
EIGEN VALUE 4.04

 

Four job characteristics variables were measured in this study:
interdependence; analyzability; variety; and discretion.

Interdgmndence. Interdependence was measured with a modified version of a
scale used by Major (1990) (which adapted three-items from Ostroff & Kozlowski
(1993)). Five items that made up the original Major (1990) scale, along with three
additional items written to be consistent with the scale, made up the eight-item scale
used in the present study. This scale measured the extent to which an individual had
to coordinate with other workgroup members to accomplish work. The rating scale for
this measure ranged from (1) very little extent to (5) very great extent. All sources
provided this measure. The scale items can be found in Appendix J.

Comlexity. Complexity was measured with modified versions of two scales

developed by Withey et a1. (1983). Their five-item scale, Exceptions, measured the

82

extent to which the newcomer’s tasks were routine and repetitious. After the factor
analysis, responses to this scale were reverse-scored to be consistent with the construct
being measured and the hypotheses developed (i.e., higher scores indicated greater
number of exceptions or variety). Their five-item scale, Analyzability, measured the
extent to which the job contained defined steps and procedures for getting work done.
The items in the two measures were rated on a scale ranging from (1) very little extent
to (5) very great extent All sources provided this measure. The items in these two
measures can be found in Appendix K.

Discretion. A modified version of West’s (1987) discretion measure was used
in this study. Three additional items were added to West’s (1987) original four-item
discretion scale to form a seven-item measure for the present study. This scale
measured the extent to which the job provided an individual with freedom to
determine how tasks were accomplished. Items on this measure were rated on a
five-point scale ranging from ( 1) very little extent to (5) very great extent. All sources
provided this measure. The scale items can be found in Appendix L.

These job characteristics are conceptually distinct. However, to test the
empirical independence of these four constructs, the items making up these four
measures were subject to a factor analysis. The factor analysis results indicated a four
factor solution, corresponding to the four job characteristics constructs. The rotated
factor matrix is presented in Table 4. The first factor contained all interdependence
items. Factor 2 contained all exceptions items, Factor 3 contained all analyzability
items, and Factor 4 included all discretion items. Each factor accounted for at least

5.5 percent of the variance.

83
Table 4

Principal Axis Factor Matrix for Job Characteristics Items

 

 

Factor

2 3 4
Interdependence8 ,& - 12 .03 - 07
Interdependence3 fl - 11 -.03 - 10
Interdependence6 10 - 16 -.11 - 07
Interdependence4 .__6_7_ - 14 .00 - 04
Interdependencel @ - 03 .09 - 09
Interdependence7 160 - 11 .08 - 16
InterdependenceS 14 10 .05 - 01
Interdependence2 112 09 -.04 - 12
ExceptionsS -.10 11 .23 -.19
Exceptions3 -.12 11 .25 -.17
Exceptions4 -.09 11 .27 -.19
Exceptions] -.09 12 .28 -.23
Exceptions2 -.07 fi .34 -. 12
Analyzability3 .03 .37 15 -.Ol
Analyzabilityl .02 .29 ,7_1 .05
Analyzability2 .02 .03 158 .02
AnalyzabilityS -.03 .30 _61 -.04
Analyzability4 .06 .24 g -.07
Discretion6 -.09 -. 14 -.04 fl
Discretion3 -.02 -.O9 .07 66
Discretion2 -.11 -.10 -.10 _._6_6_
Discretion7 -.O7 -.23 -.12 £4,
Discretion4 -.20 .04 .05 fl
Discretionl -.08 -.l l .08 fl
DiscretionS‘ -.03 -.O4 -.O3 _32
EIGEN VALUE 6.00 4.57 2.75 1.37

 

‘ This item was deleted from the final discretion scale used in the present study.

84

One item was dropped from Factor 4, discretion item 5, because it had a low
factor loading of .39. In addition, when estimates of internal consistency were run, the
item evidenced a low item-total correlation (r = .35) and a low squared multiple
correlation (R2 = .14) compared to the other items. Therefore, the final versions of the
four job characteristics scales with their reliability coefficients are as follows:
interdependence (eight items, on = .86); exceptions (five items, or = .93); analyzability
(five items, or = .85); and discretion (six items, on = .80).

Four measures of supervisor and coworker assistance were developed for this
study: supervisor information-sharing (functional); supervisor information-sharing
(evaluative); coworker information-sharing (functional); and coworker information-
sharing (evaluative).

Sum/_isor functional information-sharing. This six-item measure examined
how fiequently the supervisor provided the newcomer with information on how to
perform his/her tasks. Respondents answered these items on a five-point scale ranging
from (1) very infrequently to (5) very frequently. The newcomer provided responses
to these items. The scale items can be found in Appendix M.

Sumrvisor evaluative inforrgtion-shging. This six-item scale measured how
frequently the supervisor provided the newcomer with information on how well he/she
was performing the job. Respondents answered these items on a five-point scale
ranging from (1) very infrequently to (5) very frequently. The new employee

responded to these items. The scale items can be found in Appendix N.

85
Coworker functional information-sharing. This six-item scale measured the

extent to which coworkers provided the newcomer with information on how to
perform his/her tasks. Respondents answered these items on a five-point scale
ranging from (1) very infrequently to (5) very frequently. The new employee provided
these measures. The scale items can be found in Appendix 0.

Coworl_<_er eva_l_pa_t_ive informgtion-shariJg. This six-item scale measured the
extent to which coworkers provided the newcomer with information on how well
he/she was performing the job. Respondents answered these items on a five-point
scale ranging from (1) very infrequently to (5) very frequently. The newcomer
provided these measures. The scale items can be found in Appendix P.

The 24 items comprising these measures were subjected to a principal axis
factor analysis. The factor analysis resulted in a four-factor solution, presented in
Table 5. The factor structure basically reflected the predicted dimensions, with each
factor accounting for at least 5.2 percent of the variance.

Factor 1 included all items concerning supervisor evaluative information-
sharing, and one item concerning supervisor functional information-sharing. Factor 2
contained all items making up the coworker evaluative information-sharing measure.
Factor 3 consisted of all coworker functional information-sharing items. Factor 4
included all supervisor functional-information sharing items except the one item
loading on Factor 1. In fact, this item (supervisor functional information-sharing 3)
was actually split across Factors 1 and 4, as seen in its factor loadings presented in
Table 5. Therefore, to preserve the conceptual distinction of the two constructs, this

item was deleted from the measure used in this study.

86
Table 5

Principal Axis Factor Matrix for Supervisor and Coworker Assistance Items

 

 

Factor

1 2 3 4
Supervisor Evaluative4 ._9_2_ .18 .02 .17
Supervisor Evaluative6 ,_8_9_ .20 .04 .20
Supervisor Evaluative2 _.Q .18 .01 .18
Supervisor Evaluative3 ._81 .18 .04 .22
Supervisor EvaluativeS 12 .20 .06 .21
Supervisor Evaluativel 11 .22 -.04 .24
Supervisor Functional3‘ ._5_3_ .02 .15 .49
Coworker Evaluative4 .21 ._8_ .28 -.01
Coworker Evaluative6 .19 _._4 .26 .05
Coworker Evaluative2 .20 13 .24 .01
Coworker Evaluative3 .15 £1 .29 .04
Coworker Evaluativel . 17 1_ .25 .00
Coworker EvaluativeS .20 ._Z .30 .03
Coworker Functionall .02 .26 1Q .07
Coworker Functional6 -.01 .23 19 .18
Coworker Functiona12 -.O3 .28 ._7_4 .10
Coworker Functional4 .07 .35 11 .11
Coworker FunctionalS .05 .12 16 .18
Coworker Functional3 .09 .34 ._63 .15
Supervisor Functional6 .19 .03 .18 ._8_
Supervisor Functional2 .15 .01 .12 ,__1_
Supervisor Functionall .33 .01 .07 1_
Supervisor FunctionalS .20 .02 . 15 1;
Supervisor Functional4 .48 .05 .25 ,_9
EIGEN VALUE 9.62 4.71 2.94 1.24

 

' This item was deleted from the final supervisor functional infonnation—sharing scale
used in the present study.

87

The final measures used in the study and their reliability estimates are:
supervisor functional information-sharing (five items, a = .91); supervisor evaluative
information-sharing (six items, on = .97); coworker functional information-sharing
(six items, or = .91); coworker evaluative information-sharing (six items, a = .95).

Legdpr—membel excha_ng§. A measure of leader-member exchange developed
by Scandura and Graen (1984) was used in this study. This seven-item scale
measured the quality of the relationship that developed between the newcomer and
supervisor. It examined how the supervisor understood the needs of the newcomer
and recognized his/her potential, whether the supervisor supported the newcomer, and
whether the newcomer would defend the decisions of his/her supervisor. Newcomers
checked one of four multiple choice alternatives for each item. The scale items can be
found in Appendix Q.

The seven leader-member exchange items were subjected to a factor analysis.
One factor that included all seven items was extracted. This factor accounted for
53.0 percent of the variance. In addition, the internal consistency reliability estimate
for this scale was .85.

Team-member exchange. An adaptation of Seers’ (1989) ten-item measure of
team-member exchange quality was used in this study. It measured the willingness of
workgroup members to ask for and provide help, to switch jobs with one another, and
to let others know when they were affecting their work. Seers (1989) reported a
reliability estimate of .92 for this scale. Newcomers in the present study checked one
of four multiple choice alternatives for each item. The scale items can be found in

Appendix R.

88

The ten team-member exchange items were subject to a principal axis factor
analysis. Four factors were rotated using varimax rotation. In addition, the scree plot
suggested that four or five factors were possible. These results suggested that the
items did not measure a unidimensional construct. Rather, the four factors, comprised
of 2—3 items each, seemed to measure separate dimensions. Therefore, based on the
failure to find a unidimensional construct, it was decided that this construct would be
excluded from the analyses proposed in this study.

M_ast§g. Mastery was measured with an eight-item scale developed for this
study. It measured whether the newcomer had learned his/her tasks, had developed a
"system" for doing work, and was meeting job responsibilities. Individuals responded
to these items on a seven-point scale ranging from (1) strongly disagree to
(7) strongly agree. New employees and their coworkers answered these items. For
each newcomer, between one and three coworkers rated their mastery. Responses
from these 1-3 coworkers were averaged to form the coworker rating of the
newcomer’s mastery. The decision to average the ratings when two or three
coworkers rated one newcomer was based on the moderate consistency found between
coworker ratings. When two coworkers provided ratings for each newcomer, the
coefficient alpha for these two ratings was .43. When three coworkers provided
ratings, the coefficient alpha for the three ratings was .69. The scale items can be
found in Appendix S.

The newcomer responses to the eight mastery items were subjected to a factor
analysis. The factor analysis yielded a two-factor solution, presented in Table 6. Both

factors accounted for at least 12.5 percent of the variance. Factor 1, consisting of five

89
Table 6

Principal Axis Factor Matrix for Mastery Items

 

 

Factor
1 2
Mastery8 ._8_6 .29
Mastery7 16 .27
Mastery2 ,66 .39
Mastery3 p64 .32
Mastery4 16 .51
Mastery6‘ .34 14
Masteryl‘ .30 .11
MasteryS‘ .33 14
EIGEN VALUE 5.03 1.00

 

‘ These items were deleted from the final mastery scale used in this study.

items, appears to measure mastery of the basic requirements of the job (e.g., is
competent in carrying out assignments, completes tasks in a timely manner, and has
developed a system for performing duties). Factor 2, consisting of 3 items, measures
an absolute or final mastery of all tasks on the job. Conceptually, Factor 1 is most
consistent with the learning process examined in the present study. Because the
present study focused on the mastery of employees new to the job, the five items
comprising the first factor were selected to form the mastery measure. The internal

consistency reliability estimate for the newcomer’s perception of mastery was .89.

90
Innovation. Allen and Meyer’s (1990) modified version of Jones’ (1986)

innovation measure was used in this study. The four items of the scale, along with
three additional items written for this study, made up the seven-item innovation scale.
This scale measured the extent to which the newcomer tried to change the purpose of
his/her role, the tasks that make up the job, or the methods to accomplish his/her
tasks. Respondents answered these items on a five-point scale ranging from

(1) not at all to (5) to a great extent. Newcomers and coworkers answered these
items. Between one and three coworkers answered this scale for the newcomer in
their workgroup. The responses from the 1-3 coworkers were averaged to form the
coworker rating of the newcomer’s innovation. The decision to average the ratings
when two or three coworkers rated one newcomer was based on the moderate
consistency found between coworker ratings. When two coworkers provided ratings
for each newcomer, the coefficient alpha for these two ratings was .61. When three
coworkers provided ratings, the coefficient alpha for the three ratings was .62. The
items can be found in Appendix T.

The newcomer responses to the seven innovation items were factor analyzed,
resulting in a two-factor solution. The results of the factor analysis are presented in
Table 7. Six of the items loaded on Factor 1. Innovation item number 3, which was a
reverse-scored item, made up Factor 2. For the present study, the six items
comprising Factor 1 were used as the innovation measure, and the reverse-scored item
was dropped from further consideration. The internal consistency reliability estimate

for the newcomer’s perception of innovation was .88.

91
Table 7

Principal Axis Factor Matrix for Innovation Items

 

 

Factor

1 2
Innovation5 16 -.10
Innovation4 16 .10
Innovation7 16 .35
Innovationl ,3: .41
Innovation2 fl .29
Innovation6 12_ -.37
Innovation3‘ .02 .2_1
EIGEN VALUE 3.79 1.08

 

' This item was deleted from the final innovation scale used in this study.

Workgroup influence. This eight-item measure was developed for use in this
study. It measured the extent to which the supervisor and coworkers used the
newcomer’s advice, acted on the newcomer’s suggestions, and included the newcomer
in work-related decisions. Respondents answered these items on a seven-point rating
scale ranging from (1) strongly disagree to (7) strongly agree. Newcomers and
coworkers answered these items. For each newcomer, between one and three
coworkers provided ratings of workgroup influence. Responses from these
1-3 coworkers were averaged to form the coworker rating of the newcomer’s
influence. The decision to average the ratings when two or three coworkers rated one

newcomer was based on the moderate consistency found between coworker ratings.

92
Table 8

Principal Axis Factor Matrix for Workgroup Influence Items

 

 

Factor

1
InfluenceS .83
Influencel .78
Influence4 .76
Influence8 .76
Influence7 .75
Influence2 .71
Influence6 .69
Influence3 .67
EIGEN VALUE 4.87

 

When two coworkers provided ratings for each newcomer, the coefficient alpha for
these two ratings was .65. When three coworkers provided ratings, the coefficient
alpha for the three ratings was .52. Scale items are listed in Appendix U.

The new employee’s responses to the eight workgroup influence items were
subject to a factor analysis. One eight-item factor was extracted, accounting for
60.9 percent of the variance. The results of the factor analysis are presented in
Table 8. In addition, the internal consistency reliability estimate for the newcomer’s

perception of workgroup influence was .91.

RESULTS

The sections below describe several sets of analyses that were run to test
groups of hypotheses. Due to the nature of the analyses, hypotheses will be grouped
in a different order than they were presented in the introduction. Appendix V lists the
hypotheses relevant to each analysis that was conducted.

Patterns of Information-Seeking by Strategy, Conteng, and Timgn the Job

Hypotheses 1 through 11 examined the pattern of information-seeking used due
to strategy, content, and time on the job. Because of the number of specific
hypotheses to be tested, a 3-way (Strategy x Content x Time on the Job) repeated-
measures ANCOVA was conducted to provide some check on Type I error and an
overall description of significant effects. If the results were significant, apriori
contrasts were run to examine the specific hypotheses.

However, the results of the factor analyses for the information-seeking
strategies led to the deletion of items pertaining to observation for evaluative
information. This left only the strategy of observation for functional information
without a corresponding strategy for evaluative information. Therefore, in order to
have complete cells for strategies across content areas, the ANCOVA was run on only
three strategies (inquiry of supervisors, inquiry of coworkers, experimentation).
Hypotheses concerned with the use of observation as an information-seeking strategy
were not tested in these analyses (hypotheses 3 and 6).

Hyppthesis 1. Hypothesis I predicted a three-way interaction between strategy,
content, and time on the job to influence information-seeking frequency. To examine

the relative frequency of information-seeking across strategies, information content,

93

94

and time on the job, a repeated-measures 3 x 2 x 3 (Strategy x Content x Tenure)
ANCOVA was conducted. The two within-subjects factors were information-seeking
Strategy (experimentation, inquiry of supervisor, inquiry of coworkers) and
information Content (functional, evaluative). The between-subjects factor was time on
the job, which was trichotomized into three categories. The three Tenure categories
included: (1) less than 13 weeks; (2) 14-36 weeks; and (3) 37 or more weeks on the
job. The dependent variable was the frequency of information-seeking for each of the
strategy and content combinations. In addition, professional v. nonprofessional job
status was entered as a covariate in the ANCOVA to control for differences between
these two job categories. The job status covariate was also included in all subsequent
analyses to test specific hypotheses.

ANCOVA results are presented in Table 9. First, Mauchly’s test of sphericity
was significant for both the Strategy (x = 15.70, p < .001) and Strategy X Content
interaction (x = 11.35, p < .01), indicating that the variances of the transformed
variables were not equal. Therefore, Table 9 reports the multivariate F-tests rather
than the univariate F-tests for all effects involving Strategy (Similarly, for later
analyses testing specific hypotheses, the multivariate F-tests were reported for Strategy
effects). The job status covariate was significant (F(1, 249) = 8.94, p < .01) in this
analysis. There were also main effects for Tenure (F(2, 249) = 3.74, p < .05),
Strategy (F(2, 249) = 393.37, p < .001), and Content (F(1, 250) = 23.70, p < .001).
Significant effects were also found for a Tenure x Strategy interaction
(F(4, 498)= 2.54, p < .05), Tenure x Content interaction (F(2, 250) = 4.07, p < .05),

Strategy x Content interaction (F(2, 249) = 19.82, p < .001), and a three-way

95
Table 9

Hypothesis 1:

3 (Strategy) x 2 (Content) x 3 (Time on the Job) ANCOVA Results

 

 

Effect Numerator DF Denominator DF MS Error F'
Job Status 1 249 1.51 8.94“
Tenure 2 249 1.51 374*
Strategy 2 249 393.37“
Content 1 250 .66 23.70“
Tenure x Strategy 4 498 254*
Tenure x Content 2 250 4.07 "'
Strategy x Content 2 249 19.82"
Tenure x Strategy x Content 4 498 3.87**

 

*pS.05;**pS.01

‘F-tests for all effects involving Strategy are multivariate F-tests (MS Error was not

available for these analyses).

96

Functional Information-Seeking

 

Information-Seeking
Frequency

 

 

Tenure Tenure

1 2
Time on Job

Evaluative Information-Seeking

5..

 

 

 

 

 

 

Tenure

at 4.5 -~
:
3 4 I— _r I
3 3‘ 35
m . 4.-
;_- 5
o 3 3
2; 0'
g E 2.5 ,_
3 2 if" M
'E
— 1.5 +

1

Tenure Tenure

1 2

Time on Job

 

 

—l— Experiment
+ Ask supervisor

—°— Ask coworker

 

 

 

—l— Experiment
—*— Ask supervisor

—°— Ask coworker

 

Figure 2. Strategy x Content x Tenure Interaction.

 

 

97
Table 10

Three-Way AN COVA Results:

Cell Means and Standard Deviations

 

 

Tenure

1 2 3

Experiment Functional 3.86 3.98 3.88
(.73) (.83) (.84)

Ask Supervisor Functional 2.57 2.37 2.17
(.91) (.87) (.73)

Ask Coworker Functional 2.88 2.55 2.13
(1.10) (1.03) (.99)

Experiment Evaluative 3.98 3.92 3.88
(.71) (.79) (.78)

Ask Supervisor Evaluative 2.11 2.27 2.19
- (.85) (.99) (.91)

Ask Coworker Evaluative 2.14 2.16 1.90
(.95) (1.01) (.83)

 

98
Tenure x Strategy x Content interaction (F(4, 498) = 3.87, p < .01). Thus, these

results support the three-way interaction predicted in Hypothesis 1. Figure 2 displays
information-seeking frequency by tenure and strategy with separate plots for functional
and evaluative information. Table 10 presents the cell means and standard deviations
from the ANCOVA analysis.

To better understand the significant three-way interaction indicated in Figure 2,
an analysis of the simple interaction effects for Strategy and Content was conducted at
each level of the tenure variable. Repeated-measures 3 x 2 (Strategy x Content)
ANOVAs with the job status covariate were run for each level of tenure. These
results are presented in Table 11. The ANOVA at the first tenure level resulted in
significant main effects for Strategy (F(2, 81) = 115.94, p < .001) and Content
(F(1, 82) = 24.40, p < .001), and a significant Strategy x Content interaction
(F(2, 81) = 21.48, p < .001). The ANOVA for individuals at the second tenure level
revealed significant main effects for Strategy (F(2, 83) = 106.68, p < .001),

Content (F(1, 84) = 5.55, p < .05), and a significant Strategy x Content interaction
(F(2, 83) = 5.02, p < .01). Finally, the ANOVA for the third tenure level resulted in
significant main effects for the job status covariate (F(1, 83) = 5.93, p < .05) and
Strategy (F(2, 83) = 188.47, p < .001). The main effect for Content (F(1, 84) = 1.07,
p > .05) and the Strategy x Content interaction (F(2, 83) = 2.56, p > .05) were not
statistically significant.

Individuals generally sought information to a different degree across the three
strategies for all three tenure levels. For the first two tenure levels, the Content x

Strategy interaction suggests that the use of information-seeking strategies generally

Understanding the Three-Way Interaction:

99
Table 11

Test of Simple Interaction Effects at Each Tenure Level

A. Tenure Level = 1

 

 

Effect Numerator DF Denominator DF MS Error F'
Job Status 1 81 1.35 1.48
Strategy 2 81 115.94“
Content 1 82 .66 24.40M
Strategy x Content 2 81 21.48"

 

B. Tenure level = 2

 

 

Effect Numerator DF Denominator DF MS Error F‘
Job Status 1 83 1.73 1.90
Strategy 2 83 106.68"
Content 1 84 .75 555*
Strategy x Content 2 83 5.02"

 

C. Tenure Level = 3

 

 

Effect Numerator DF Denominator DF MS Error F‘
Job Status 1 83 1.46 593*
Strategy 2 83 188.47“
Content 1 84 .57 1.07
Strategy x Content 2 83 2.56

 

*ps.05;**ps.01

'F-tests for all effects involving Strategy are multivariate F-tests (MS Error was not
available for these analyses).

100
depended on the content of the information being sought For the third tenure level,

information content did not impact the use of information-seeking strategies.

A second set of analyses were conducted to examine the simple interaction
effects for Strategy and Tenure for the two types of information content. Repeated-
measures 3 x 3 (Strategy x Tenure) ANOVAs with the job status covariate were run
separately for functional and evaluative information. The results of these analyses are
presented in Table 12. The AN OVA for functional information resulted in significant
main effects for the job status covariate (F(1, 249) = 6.67, p S .01), Tenure
(F(2, 249) = 8.19, p < .001) and Strategy (F(2, 249) = 210.92, p < .001), and a
significant Tenure x Strategy interaction (F(4, 498) = 3.62, p < .01). The ANOVA for
evaluative information revealed significant main effects for the job status covariate
(F(1, 249) = 5.90, p < .05) and Strategy (F(2, 249) = 416.30, p < .001). The Tenure
main effect (F(2, 249) = .74, p > .05), and Tenure x Strategy interaction
(F(4, 498) = 1.80, p > .05) were not statistically significant

These analyses indicate that, for functional information, there is a change in
general in the use of information-seeking strategies with different levels of tenure.
However, for evaluative information, while there is a difference in the use of
information-seeking strategies overall (Strategy main effect), tenure on the job does

not affect the relative use of information-seeking strategies.

101
Table 12

Understanding the 3-Way Interaction:

Tests of Simple Interaction Effects for Each Information Content Category

A. Content = Functional Information

 

 

 

 

 

Effect N umerator DF Denominator DF MS Error F‘
Job Status 1 249 .90 6.67“
Tenure 2 249 .90 8. 19**
Strategy 2 249 210.92”
Tenure x Strategy 4 498 3.62“
B. Content = Evaluative Information

Effect Numerator DF Denominator DF MS Error F‘
Job Status 1 249 1.27 5.90*
Tenure 2 249 1.27 .74
Strategy 2 249 416.30"
Tenure x Strategy 4 498 1.80

 

*ps.05;**pS.01

'F-tests for all effects involving Strategy are multivariate F-tests (MS Error was not
available for these analyses).

102
Hypotheses 2 and 4. In addition to these overall analyses, specific analyses

were run to test hypotheses 2, 4, 5, and 7-11. Hypotheses 2 and 4 concerned the
relative use of information-seeking strategies to acquire functional information,
predicting that experimentation would be used most frequently, followed by inquiry of
coworkers and inquiry of supervisors. As explained earlier, observation for functional
information (hypothesis 3) was not examined in the ANOVA analyses.

To test these hypotheses, a one-way ANOVA with 3 strategy levels and the job
status covariate was conducted. Two apriori contrasts were specified:
(1) experimentation was compared to inquiry of coworkers and inquiry of supervisors
for functional information; and (2) inquiry of coworkers was compared to inquiry of
supervisors for functional information. Table 13 presents the ANOVA results.

Results indicated a significant main effect for the job status covariate
(F(1, 251) = 9.83, p < .01) and Strategy (F(2, 251) = 205.96, p < .001). The mean
frequency for each of the three information-seeking strategies were as follows:
experimentation (x = 3.91, sd = .80); inquiry of coworkers (x = 2.52, sd = 1.08); and
inquiry of supervisors (x = 2.37, sd = .86). Results for the specified contrasts
indicated that experimentation was significantly different than inquiry of coworkers
and inquiry of supervisors for functional information (F(1, 252) = 393.05, p < .001),
and that inquiry of coworkers was significantly different from inquiry of supervisors
for functional information (F(1, 252) = 4.51, p < .05). These results indicate that
individuals acquire functional information most frequently through experimentation,
followed by inquiry of coworkers, and finally inquiry of supervisors. These results

support hypotheses 2 and 4.

103
Table 13

Hypotheses 2 and 4: AN OVA Results for Functional Information-Seeking

 

 

Effect Numerator DF Denominator DF MS Error F‘
Job Status 1 251 .95 9.83“
Strategy 2 251 205.96"
Exprmt v. Ask Cowk & Ask Supv 1 252 .92 393.05”
Ask Cowk v. Ask Supv l 252 .63 4.51*

 

*pS.05;**pS.Ol

'F-test for the Strategy effect is a multivariate F-test (MS Error was not available for
this analysis).

However, the ANOVAs for the simple interaction effects of Strategy and
Tenure that were run separately for functional and evaluative information indicated a
significant Strategy X Tenure interaction for functional information. Therefore, to
understand the relative frequency of information-seeking strategies for different levels
of tirrre on the job, the one-way ANOVA, with 3 strategy levels and the contrasts
specified above, was run separately for each tenure level. These results are
summarized in Table 14. For the first tenure level, the AN OVA resulted in a
significant main effect for Strategy (F(2, 81) = 40.42, p < .001). In addition, the
apriori contrasts indicated that experimentation was significantly different from inquiry
of coworkers and inquiry of supervisors (F(1, 82) = 78.17, p < .001), and that inquiry
of supervisors was significantly different from inquiry of coworkers

(F(1, 82) = 4.47, p < .05).

104
Table 14

Hypotheses 2 and 4: AN OVA Results for Functional Information-Seeking
at Each Tenure Level

A. Tenure Level = l

 

 

 

 

 

 

 

 

Efiect Numerator DF Denominator DF MS Error F‘
Job Status 1 81 .78 .17
Strategy 2 81 40.42"
Exprmt v. Ask Cowk & Ask Supv 1 82 .92 78.17M
Ask Cowk v. Ask Supv 1 82 .89 4.47*
B. Tenure Level = 2

Efi’ect Numerator DF Denominator DF MS Error F‘
Job Status 1 83 .96 2.70
Strategy 2 83 68.60"
Exprmt v. Ask Cowk & Ask Supv l 84 1.06 122.99"
Ask Cowk v. Ask Supv 1 84 .47 3.09
C. Tenure Level = 3

Effect N umerator DF Denominator DF MS Error F‘
Job Status 1 83 .97 400*
Strategy 2 83 124.59“
Exprmt v. Ask Cowk & Ask Supv l 84 .69 247.25"
Ask Cowk v. Ask Supv 1 84 .52 .14

 

*pS.05;**pS.01

‘F-tests for Strategy effects are multivariate F-tests (MS Error was not available for
these analyses).

105

For the second tenure level, ANOVA results indicated a significant main effect
for Strategy (F(2, 83) = 68.60, p < .001). The contrast results demonstrated that
experimentation was significantly different from inquiry of coworkers and inquiry of
supervisors for functional information (F(1, 84) = 122.99, p < .001). However, the
contrast between inquiry of coworkers and inquiry of supervisors was not significant
(F(1, 84) = 3.09, p > .05).

Finally, for the third tenure level, significant main effects were found for the
job status covariate (F(1, 83) = 4.00, p < .05) and Strategy (F(2, 83) = 124.59,

p < .001). In addition, the contrast comparing experimentation with inquiry of
coworkers and inquiry of supervisors for functional information was significant
(F(1, 84) = 247.25, p < .001). The contrast comparing inquiry of coworkers and
inquiry of supervisors for functional information was not significant

(F(1, 84) = .14, p > .05).

In sum, these results suggest that, overall, individuals acquire functional
information most frequently through experimentation compared to inquiry of
coworkers and supervisors. In addition, individuals on the job for 13 weeks or less
ask their coworkers more fi'equently than their supervisors for functional information.
Although significant, the difference between asking coworkers and asking supervisors
for functional information is small. Further, for those on the job for more than 13
weeks, there is no difference in the frequency with which they ask coworkers and
supervisors for functional information. The specific nature of the effects of tenure on

information-seeking will be clarified in the tests for hypotheses 8 and 9.

106
Hypotheses 5 and 7. Hypotheses 5 and 7 predicted that individuals would

acquire evaluative information most frequently through experimentation, followed by
inquiry of supervisors, and inquiry of coworkers. To test these hypotheses, a one-way
ANOVA with 3 strategy levels and the job status covariate was conducted. Two
apriori contrasts were specified: (1) experimentation was compared to inquiry of
supervisors and inquiry of coworkers for evaluative information; and (2) inquiry of
supervisors was compared to inquiry of coworkers for evaluative information. The
results for this AN OVA can be found in Table 15.

The ANOVA resulted in significant main effects for the job status covariate
(F(1, 251) = 6.16, p < .05) and Strategy (F(2, 251) = 417.58, p < .001). The means
and standard deviations for the three information-seeking strategies to acquire
evaluative information were as follows: experimentation (x = 3.93, sd = .76); inquiry
of supervisor (x = 2.19, sd = .92); and inquiry of coworkers (x = 2.07, sd = .94).
Results of the apriori contrasts indicated that experimentation was significantly
different from inquiry of supervisors and coworkers for evaluative information
(F(1, 252) =.830.89, p < .001), and that inquiry of supervisors was significantly
different from inquiry of coworkers for evaluative information (F(1, 252) = 5.59,
p < .05). These results indicate that individuals rely most frequently on
experimentation to acquire evaluative information, followed by inquiry of supervisors
and inquiry of coworkers. These results support hypotheses 5 and 7. However, it
should be noted that the difference between asking supervisors and asking coworkers,

while significant, is very small.

107
Table 15

Hypotheses 5 and 7: AN OVA Results for Evaluative Information-Seeking

 

 

Effect Numerator DF Denominator DF MS Error F‘
Job Status 1 251 1.26 6.16*
Strategy 2 251 417.58”
Exprmt v. Ask Supv & Ask Cowk 1 252 .66 830.89“
Ask Supv v. Ask Cowk 1 252 .35 5.59“

 

*ps.05;**ps.01

‘F-test for the Strategy effect is a multivariate F-test (MS Error was not available for
this analysis).

Hypothesis 8. Hypothesis 8 predicted that inquiry of supervisors for functional
information would decrease with increased time on the job. To test this hypothesis, a
one-way ANOVA was run. The dependent variable was inquiry of supervisor for
functional information. The job status covariate was entered first, and the independent
variable was tenure. Two apriori contrasts were specified: (1) inquiry of supervisors
for tenure level 1 was compared to inquiry of supervisors for tenure level 2; and
(2) inquiry of supervisors for tenure level 2 was compared to inquiry of supervisors
for tenure level 3. AN OVA results are presented in Table 16.

AN OVA results indicated a significant main effect for Tenure
(F(2, 249) = 3.95, p < .05). The mean frequency for inquiry of supervisors for
functional information at each tenure level were as follows: (1) tenure level 1
(x = 2.57, sd = .91); (2) tenure level 2 (x = 2.37, sd = .87); and (3) tenure level 3

(x = 2.17, sd = .73). The apriori contrasts indicated that inquiry of supervisors for

108

functional information was significantly greater for individuals at tenure level 1 than
for individuals at tenure level 2 (F(1, 249) = 5.65, p < .05).

However, individuals at tenure levels 2 and 3 did not significantly differ in
their use of inquiry of supervisors for functional information (F(1, 249) = 2.28,

p > .05). These results provide partial support for hypothesis 8.

Table 16
Hypothesis 8: AN OVA Results for Inquiry of Supervisor

for Functional Information

 

 

Effect SS DF MS F
Within Cells 176.01 249 .71

Job Status 1.34 1 1.34 1.90
Tenure 5.58 2 2.79 3.95*
Tenure 1 v. Tenure 2 4.00 1 4.00 5.65*
Tenure 2 v. Tenure 3 1.61 1 1.61 2.28

 

*pS.05;**pS.Ol

Hypothesis 9. Hypothesis 9 predicted that inquiry of coworkers for functional
information would decrease with increased time on the job. A one-way AN OVA with
three tenure levels and the job status covariate was run with inquiry of coworkers for
functional information as the dependent variable. Two apriori contrasts were
specified: (1) inquiry of coworkers for tenure level 1 was compared to inquiry of
coworkers for tenure level 2; and (2) inquiry of coworkers for tenure level 2 was

compared to inquiry of coworkers for tenure level 3. Results are reported in Table 17.

109
Table 17

Hypothesis 9: AN OVA Results for Inquiry of Coworkers

for Functional Information

 

 

Effect SS DF MS F
Within Cells 260.37 249 1.05

Job Status 9.91 l 9.91 9.48**
Tenure 18.74 2 9.37 8.96**
Tenure l v. Tenure 2 11.76 1 11.76 11.25**
Tenure 2 v. Tenure 3 7.09 1 7.09 6.78“

 

*pS.05;**pS.01

ANOVA results revealed a main effect for the job status covariate
(F(1, 249) = 9.48, p < .001) and for time on the job (F(2, 249) = 8.96, p S .001). The
mean frequency for inquiry of coworkers for functional information at each tenure
level were as follows: (1) tenure level 1 (x = 2.88, sd = 1.10); (2) tenure level 2
(x = 2.55, sd = 1.03); and (3) tenure level 3 (x = 2.13, sd = .99). The apriori contrasts
indicated that inquiry of coworkers for functional information was significantly greater
at tenure level 1 compared to tenure level 2 (F(1, 249) = 11.25, p < .001). In
addition, inquiry of coworkers for functional information was significantly greater at
tenure level 2 compared to tenure level 3 (F(1, 249) = 6.78, p S .01). These results
support hypothesis 9.

Hypothesis 10. Hypothesis 10 predicted that inquiry of supervisors for
evaluative information would increase with greater time on the job. A one-way

ANOVA with three tenure levels and the job status covariate was run. Two apriori

110

contrasts were specified: (1) inquiry of supervisors for evaluative information for
tenure level 1 would be less than inquiry of supervisors for tenure level 2; and

(2) inquiry of supervisors for tenure level 2 would be less than inquiry of supervisors
for tenure level 3. AN OVA results are displayed in Table 18.

Results of the ANOVA indicated a main effect for the job status covariate
(F(1, 249) = 4.21, p < .05). The effects for Tenure and the apriori contrasts were not
significant. These results did not support hypothesis 10.

Hypothesis 11. Hypothesis 11 predicted that inquiry of coworkers for
evaluative information would increase with greater time on the job. A one-way
ANOVA with three tenure levels and the job status covariate was run. Two apriori
contrasts were specified: (1) inquiry of coworkers for evaluative information for
tenure level 1 would be less than inquiry of coworkers for tenure level 2; and
(2) inquiry of coworkers for tenure level 2 would be less inquiry of coworkers for
tenure level 3. ANOVA results are reported in Table 19.

Results of the ANOVA indicated a main effect for the job status covariate
(F(1, 249) = 10.07, p < .001). The effects for Tenure and the apriori contrasts were

not significant. These results did not support hypothesis 11.

111
Table 18

Hypothesis 10: AN OVA Results for Inquiry of Supervisor

for Evaluative Information

 

 

Effect SS DF MS F
Within Cells 207.78 249 .83

Job Status 3.52 1 3.52 4.21*
Tenure 1.72 2 .86 1.03
Tenure 1 v. Tenure 2 1.49 l 1.49 1.78
Tenure 2 v. Tenure 3 .23 1 .23 .27

 

*ps.05;**ps.01

Table 19
Hypothesis 11: AN OVA Results for Inquiry of Coworkers

for Evaluative Information

 

 

Effect SS DF MS F
Within Cells 208.39 249 .84

Job Status 8.43 1 8.43 10.07"
Tenure 2.59 2 1.29 1.54
Tenure 1 v. Tenure 2 .07 1 .07 .09
Tenure 2 v. Tenure 3 2.52 1 2.52 3.01

 

*pS.05;**pS.01

112

The remaining analyses focused on antecedent factors and how they related to
information-seeking strategies, and how information-seeking strategies relate to
socialization outcomes. The means, standard deviations, and intercorrelations for the
variables examined in this study are presented in Table 20. Intercorrelations between
the information-seeking strategies ranged from -.05 to .62. Correlations were higher
within an information strategy (e.g., between inquiry of coworkers for functional and
evaluative information). In addition, inquiry of coworkers for functional and
evaluative information were more highly correlated with observation for functional
information than the other strategies. Although these strategies were correlated, they
were conceptualized to be theoretically distinct. Furthermore, the factor analysis did
result in seven distinct factors. It was felt that new employees could distinguish
between information-seeking strategies and information content.

The intercorrelation matrix also reveals high correlations between measures of
supervisor and coworker sharing of functional and evaluative information, and the
respective scales measuring new employee information-seeking from supervisors and
coworkers for the two types of information. The high intercorrelations may indicate a
shared climate, or may be the result of method bias. However, the present study did
hypothesize a positive relationship between the constructs of information-sharing and
information-seeking. In addition, the inquiry of supervisor and coworkers items and
the supervisor and information-sharing items were asked in separate sections of the
survey, with explicit directions for answering each. Therefore, it was felt that new
employees could most likely distinguish their attempts to seek information from

supervisor and coworkers attempts to approach them and provide information.

113

In addition, it may be argued that observation for functional information and
coworker functional information-sharing are similar constructs. However, the present
study conceptualized coworker functional information-sharing as verbal assistance,
such as coworkers telling or describing to newcomers how to perform their jobs.
While coworkers do this, they may also provide opportunities for newcomers to
observe their behavior by actually demonstrating how to perform tasks. However,
observation for functional information was also conceptualized to include instances
where newcomers were not interacting with coworkers, but unobtrusively monitoring
others’ behaviors in order to learn their jobs. To ascertain whether these were indeed
separate constructs, a principle axis factor analysis with varimax rotation was
conducted. It resulted in two clearly interpretable factors corresponding to the two
constructs conceptualized in this study. Factor analysis results are presented in
Appendix W. Observation for functional information and coworker functional

information-sharing were therefore treated as separate constructs in study analyses.

114

Table 20

Means, Standard Deviations, and Intercorrelations

 

AskSupF
AskakF
ObserveF
13th
Ask SupE

Expermt E
Tenure

. Job Status
10. Self-eff
11. Interdep
12. Exceptns
13. Analyzblty
14. Discretion
15. LMX

16. Sup Shr F
17. ka Shr F
18. Sup Shr E
19. CW]: Shr E
20. MastrSlf
21. Innov-Slf
22. Infl-Slf
23. Mastr-ka
24. Innov-ka
25. Infl-ka

wesovswwr

M
SD

(AskakE.

-.03 .02

.24 .26 -.06

.52 .55 -.01 .60

.01 .19 .31 .18 .21
-.28 -. 17 .01 .04 -. 10 -.05

.23 .19 -.01 . 12 .21 .01 -.16
-.24 -.09 .10 -.03 .01 .29 .06

.31 .39 -. 14 .19 .32 .13 -.09

.21 . 17 .10 . 15 . 16 .02 -.08
-.13 -.01 -. 10 .02 -.03 .14 -.07
-. ll -.06 .35 -.01 -.06 .17 .09
-.06 -.05 .03 .07 -.03 .11 .(I)

.25 .14 -.17 .29 .14 -.09 -. 16

.70 .60 -.02 . 19 .51 .04 -.28

.03 . 12 .(X) .44 .26 . 15 .06

.32 .39 .02 .42 .67 .25 -.06
-.23 -.05 .14 . ll .09 .26 .22
-.17 -.01 .25 . 12 .09 .15 .27
-.08 .04 .19 .18 .13 .24 .09
-.19 -.08 .07 .03 .03 -.02 .12
-. 15 -.07 .15 .09 .06 .04 .22
-. 17 -.11 -.02 .07 .03 -.06 .20

2.52 3.11 3.91 2.19 2.07 3.93 2.01
1.08 1.11) 0.80 0.92 0.94 0.76 0.82

9 10 11 12 13 14

-.27
.23 -.09
.51 -.15 .16

-.36 .31 -.03 -.56
.10 .15 -.23 .25 -.01
-.09 .16 -.06 .02 .26 .23
.07 -.06 .13 .05 .06 -. 12
. 12 -. 12 .28 .11 -.03 -. 10
-.01 .10 .04 .01 .20 .10
-.08 .12 .24 .01 .09 -.(YZ
-.26 .55 -.07 -.22 .34 .17
.02 .10 . 12 . 15 -. 13 .23
.07 .20 -.01 . 16 .09 .28
-.11 .09 -.11 -.05 .01 .06

:14 :06 .06 :19 £08 :15

0.83 4.44 3.23 3.72 3.03 3.76
0.38 0.60 0.80 l.(X) 0.91 0.72

1 15
Table 20 (cont’d.)

 

1516171819202122232425
15.LMX --
16. Sup ShrF .32
11ka ShrF -.06 .32
18. Sup ShrE .55 .46 .12
r9.ka ShrE .06 .15 .56 .43
20. Mastr-Slf .25 -.04 -.11 .25 .25
21.1nnov-Slf -.os -.24 -.14 .11 .15 .16
22.1nn-srr .51 .09 -.03 .40 .16 .35 .21
23.Mastr-ka -.os -.13 -.12 -.02 .10 .18 .06 .12
24.1nnov-ka -.17 -.18 -.15 -.m .01 .15 .38 .13 .28
25.1nfl-ka .04 -.10 -.17 .06 -.02 .05 .16 .27 .40 .38

M 3.04 2.61 3.(X) 2.84 2.63 5.93 2.67 5.30 5.99 2.48 5.08
SD 0.54 1.01 1.00 1.16 1.08 0.90 0.92 0.91 0.86 0.84 1.05

 

Nptg. Ask Sup F = Inquiry of supervisor - functional information; Ask ka F = Inquiry of coworkers -
functional information; Observe F = Observation - functioml information; Expermt F = Experimentation
- functional information; Ask Sup E = Inquiry of supervisor - evaluative information; Ask ka E =
Inquiry of coworkers - evaluative information; Expermt E = Experimentation - evaluative information;
Tenure = time on the job; Job Status = Job status (nonpofessional/professional); Self-eff = Self-
efficacy; Interdep = Interdependence; Exceptrrs = Exceptions; Analyzblty = Analyzability; Discretion =
Discretion; LMX = Leader-member exchange; Sup Shr F = Supervisor functional information-sharing;
ka Shr F = Coworker functional information-sharing; Sup Shr E = Supervisor evaluative infonnation-
sharing; ka Shr E = Coworker evaluative information-sharing; Mastr-Slf = Mastery - self rating;
Innov—Slf = Innovation - self rating; Infl-Slf = Workgroup influence - self rating; Mastr-ka =

Mastery - coworker rating; Innov-ka = Innovation - coworker rating; Infl-ka = Workgroup influence
- coworker rating.

r > .12, p < .05; r > .16, p < .01; n = 253.

1 16
Job Characteristics and Self-Efficacy as Antecedents to Information-Seeking

Hypotheses 12 through 43 examined the relationships between various job
characteristics and the information-seeking strategies used by newcomers. In addition,
these relationships were hypothesized to be moderated by self-efficacy (hypotheses 12a
through 43a). However, because observation for evaluative information was dropped
as a variable in this study, hypotheses pertaining to this construct could not be tested
(hypotheses 18, 18a, 26, 26a, 34, 34a, 42, and 42a). Therefore, seven separate
moderated hierarchical regression analyses were run, one for each of the remaining
information-seeking strategies.

A similar procedure was followed for each information-seeking strategy. First,
the job status variable was not a covariate in these analyses. As can be expected, and
as shown in the discriminant analysis (see Table 1), the difierence between
professional and nonprofessional jobs is largely due to differences in job
characteristics. Therefore, it was not appropriate to control for job status when the
purpose of the analyses was to assess the effect of job characteristics on information-
seeking.

However, for two of the analyses, time on the job was entered first into the
regression equation as a covariate. Because the ANOVA results presented earlier
indicated that time on the job had a significant effect on two of the strategies, the
tenure variable was entered as a covariate for the following dependent variables:

(1) inquiry of supervisors for functional information; and (2) inquiry of coworkers for
functional information. For the remaining five analyses, time on the job was not

included as a covariate. In the second step of each regression, the four job

117

characteristics and self-efficacy were entered to test for their linear effects on the
information-seeking strategy. In step 3, the interaction between each job characteristic
and self-efficacy were entered to test for the role of self-efficacy as a moderator
variable.

Hymtheses 12, 123, 20, 209.. 28, 2891, 36, and 36a. These hypotheses examined
how job characteristics and self-efficacy were related to inquiry of supervisors for
functional information. Results of the moderated hierarchical regression analysis to
test these hypotheses are presented in Table 21. The first step in the equation, the
covariate, accounted for a significant amount of variance in inquiry of supervisors for
functional information. Specifically, the tenure covariate was negatively related to the
use of inquiry of supervisors for functional information (B = -.l9). The second step in
the equation accounted for a significant amount of variance in the dependent variable
above that accounted for by the covariate. The specific variables contributing to this
step were discretion (B = -.25) and self-efficacy (B = -.14), which were both
negatively related to inquiry of supervisors for functional information. These results
supported hypothesis 36. The third step in the regression equation, containing the
interactions between job characteristics and self-efficacy, was not significant. Thus,
hypotheses 12, 12a, 20, 20a, 28, 28a, and 36a were not supported by the results,
although self-efficacy was related to inquiry of supervisors for functional information

in the predicted direction.

118
Table 21

Moderated Hierarchical Regression Results

Dependent Variable: Inquiry of Supervisor for Functional Information

 

 

Step Predictors B R2 AR2
1 Tenure -.19** .O4** .O4**
2 Self-Efficacy -. 14*

Interdependence .07

Exceptions . 13

Analyzability .05

Discretion -.25** .14** .10**

3 Interdependence x Self-Efficacy 1.40**

Exceptions x Self-Efficacy -.62

Analyzability x Self-Efficacy -.07

Discretion x Self-Efficacy .44 .17“ .03‘

 

*ps.05;**ps.01

'pS.lO

119
Hypptheses 13, 13a, 21, 21%. 29, 29a, 37, and 37a. These hypotheses examined

how job characteristics and self-efficacy were related to inquiry of coworkers for
functional information. The results of the moderated hierarchical regression testing
these hypotheses are presented in Table 22. The first step in the regression equation,
tenure, was significant (B = -.28). The second step in the regression equation was also
significant. Specifically, self-efficacy had a negative relationship (B = -.18) and
interdependence had a positive relationship (B = .25) with inquiry of coworkers for
functional information. These results provided support for hypothesis 13. The third
step in the equation, consisting of the interaction terms, did not reach significance.
Thus, hypotheses 13a, 21, 21a, 29, 29a, 37, and 37a were not supported by the results,
although self-efficacy was related to inquiry of coworkers for functional information in
the predicted direction.

Hypotheses 14, 14a, 22, 224, 30, 306, 38, and 38a. These hypotheses examined
how job characteristics and self—efficacy were related to observation for functional
information. The results of this regression are presented in Table 23. The first step in
the equation accounted for a significant amount of variance in the dependent variable.
Specifically, interdependence (B = .36) and exceptions (B = .17) were positively
related to observation for functional information . These results were opposite to
those predicted by hypotheses 14 and 22. The second step in the equation was not
significant, indicating that job characteristics and self—efficacy do not interact to
predict observation for functional information. These results do not support

hypotheses 14a, 22a, 30, 30a, 38, and 38a.

120
Table 22

Moderated Hierarchical Regression Results

Dependent Variable: Inquiry of Coworkers for Functional Information

 

 

Step Predictors B R2 AR2
1 Tenure -.28** .O8** .08**
2 Self-Efficacy -.18**

Interdependence .25**

Exceptions .13

Analyzability -.01

Discretion -.O4 .22** .14**
3 Interdependence x Self-Efficacy -.66

Exceptions x Self-Efficacy -.11

Analyzability x Self-Efficacy .14

Discretion x Self-Efficacy -1.30* .25** .03‘

 

*ps.05;**ps.01

'pS.10

121
Table 23

Moderated Hierarchical Regression Results

Dependent Variable: Observation for Functional Information

 

 

Step Predictors B R2 AR2
1 Self-Efficacy -.O7

Interdependence .36"

Exceptions . 17*

Analyzability .12

Discretion -.Ol .18" .18**
2 Interdependence x Self-Efficacy -.38

Exceptions x Self-Efficacy .46

Analyzability x Self-Efficacy -. 14

Discretion x Self-Efficacy -.95' .19“ .01

 

*pS.05;**pS.01

‘ps.10

122
Hypotheses 15, 15a, 23, 23;, 31, 312: 39, and 39a. These hypotheses

concerned how job characteristics and self-efficacy were related to experimentation for
functional information. The results of the moderated hierarchical regression are
presented in Table 24. The first step in the equation, examining the linear effects of
job characteristics and self-efficacy on the dependent variable, was significant.
Discretion was positively related to experimentation for functional information

(B = .33), as predicted by hypothesis 39. The second step in the regression,
containing the interaction terms, was not significant. These results did not support
hypotheses 15, 15a, 23, 23a, 31, 31a, and 39a.

Hypptheses 16, 16a, 24, 24a, 32, 32;, 40, and 40a. These hypotheses examined
how job characteristics and self-efficacy were related to inquiry of supervisors for
evaluative information. The results of this regression equation are presented in
Table 25. The first step in the equation was statistically significant. Specifically,
interdependence was positively related to inquiry of supervisors for evaluative
information (B = .15), as predicted by hypothesis 16. Analyzability was also
positively related to inquiry of supervisors for evaluative information (B = .16), in
opposition to the prediction of hypothesis 32. Finally, exceptions was positively
related to inquiry of supervisors for evaluative information (B = .22), in support of
hypothesis 24. The second step in the equation, containing the interaction terms, was
not significant. These results do not support hypotheses 16a, 24a, 32, 32a, 40, and
40a.

123
Table 24

Moderated Hierarchical Regression Results

Dependent Variable: Experimentation for Functional Information

 

 

Step Predictors B R2 AR2
1 Self-Efficacy .09

Interdependence -.06

Exceptions -.O4

Analyzability -.14‘

Discretion .33** .14** .14“
2 Interdependence x Self-Efficacy -.27

Exceptions x Self-Efficacy 1.92“

Analyzability x Self-Efficacy 1.55*

Discretion x Self-Efficacy -.84 .17" .03'

 

*pSDi**pSDl

‘pS.10

124
Table 25

Moderated Hierarchical Regression Results

Dependent Variable: Inquiry of Supervisor for Evaluative Information

 

 

Step Predictors B R2 AR2
1 Self-Efficacy -.03

Interdependence .15“

Exceptions .22*

Analyzability .16*

Discretion -.03 .07** .O7*
2 Interdependence x Self-Efficacy .51

Exceptions x Self-Efficacy .27

Analyzability x Self-Efficacy -.56

Discretion x Self-Efficacy -.96 .09” .02

 

*pS.05;**pS.01

‘pS.10

125
It is surprising that exceptions and analyzability had different relationships with

inquiry of supervisors for evaluative information, since they were conceptualized as
two dimensions of job complexity. However, there is some indication that
analyzability acted as a suppressor variable. First, it was highly intercorrelated with
exceptions (r = -.56). Further, while its intercorrelation with inquiry of supervisors for
evaluative information was almost nonexistent (r = .02), its beta-weight when entered
into the regression was much stronger (B = .16). The beta-weight for exceptions also
increased compared to its correlation with inquiry of supervisors for evaluative
information.

Hypo—theses 17Ll7a, 25, 25;, 33, 33a, 41, and 41a. These hypotheses focused
on how job characteristics and self-efficacy were related to inquiry of coworkers for
evaluative information. The results of this moderated hierarchical regression are
presented in Table 26. The first step in the equation explained a significant amount of
variance in the dependent variable. Interdependence was positively related to inquiry
of coworkers for evaluative information (B = .29), as predicted by hypothesis 17.
Exceptions (B = .16) was also positively related to the dependent variable, supporting
hypothesis 25. The second step in the equation was not statistically significant. These
results do not support hypotheses 17a, 25a, 33, 33a, 41, and 41a.

Hypotheses 19, 194, 27, 27a, 35, 356, 43, and 43a. These hypotheses examined
how job characteristics and self-efficacy were related to experimentation for evaluative

information. The results of this regression equation are presented in Table 27.

126
Table 26

Moderated Hierarchical Regression Results

Dependent Variable: Inquiry of Coworkers for Evaluative Information

 

 

Step Predictors B R2 AR2
1 Self-Efficacy .05

Interdependence .29”

Exceptions . 16*

Analyzability .05

Discretion -.04 .12** .12**
2 Interdependence x Self-Efficacy -.16

Exceptions x Self-Efficacy -.06

Analyzability x Self-Efficacy -1.09

Discretion x Self-Efficacy -.53 .13“ .01

 

*pS.05;**pS.Ol

127
Table 27

Moderated Hierarchical Regression Results

Dependent Variable: Experimentation for Evaluative Information

 

 

Step Predictors B R2 AR2

1 Self-Efficacy .26*"'

Interdependence .19**

Exceptions .04

Analyzability .09

Discretion .16* .14** .14**
2 Interdependence x Self-Efficacy -.68

Exceptions x Self-Efficacy .37

Analyzability x Self-Efficacy .59

Discretion x Self-Efficacy -.13 .15" .01

 

*pS.05;**pS.01

128

The first step in the equation accounted for a significant amount of variance in the
dependent variable. Self-efficacy was positively related to experimentation for
evaluative information (B = .26), in an opposite direction to that hypothesized.
Discretion was positively related to experimentation for evaluative information
(B = .16), as predicted in hypothesis 43. Finally, interdependence was also positively
related to experimentation for evaluative information (B = .19), in an opposite
direction to that predicted by hypothesis 19. The second step in the regression
equation did not reach statistical significance. These results do not support hypotheses
19, 19a, 27, 27a, 35, 35a, and 43a.
Workggoup Relations and Information-Seeking

Hypotheses 45 through 53 examined how supervisor and coworker information-
sharing, the leader member-exchange (LMX) relationship, and the team-member
exchange (TMX) relationship, related to the newcomer’s willingness to seek
information through all four information-seeking strategies. Specifically, it was
hypothesized that these constructs would have a positive relationship with the
newcomer’s information-seeking attempts. However, three hypotheses (50, 54, and
55) could not be tested because TMX and observation for evaluative information were
dropped as constructs in the present analyses. For the remaining hypotheses, seven
separate regression analyses were conducted, one for each of the remaining seven
information-seeking strategies.

In the first step of the equation, the covariate(s) were entered. Job status was
used as a covariate in all seven analyses. In addition, tenure was used as a covariate

for two strategies as in the previous regression analyses: inquiry of supervisors for

129

functional information; and inquiry of coworkers for functional information. In the
second step of the equation, the variables of interest were entered.

Hypotheses 44 and 52. Hypotheses 44 and 52 predicted that supervisor and
coworker information-sharing, and leader-member exchange, would be positively
related to inquiry of supervisors for functional information. Results of the regression
analysis are presented in Table 28. The first step of the equation, the covariates, was
statistically significant. The tenure covariate was negatively related to the dependent
variable (B = -.18). The second step in the equation accounted for a significant
amount of variance in the dependent variable beyond that accounted for by the
covariates. The results indicated that supervisor information-sharing was positively
related to inquiry of supervisors for functional information (B = .69). In addition, the
results indicated that LMX was negatively related to inquiry of supervisors for
functional information (B = -.10). However, the beta weight for this variable had an
opposite sign compared to the correlation between LMX and inquiry of supervisors for
functional information (r = .12). It is also clear from the intercorrelations between the
predictor variables (e.g., see Table 20), that LMX was significantly related to
supervisor information-sharing. Therefore, LMX was acting as a suppressor variable
in this analysis. These results provide partial support for hypothesis 44; only
supervisor information-sharing was found to be positively related to inquiry of
supervisors for functional information. The suppressor effect found for LMX did not

support hypothesis 52.

130
Table 28

Moderated Hierarchical Regression Results

Supervisor and Coworker Information-Sharing, and Leader-Member Exchange (LMX)

 

 

 

 

 

Dependent Variable Step Predictor B R2 AR2
Inquiry Supervisor 1 Tenure -. 18""'I
Functional Job Status .09 .04** .04**
2 Supervisor Info-Sharing‘ .69“
Coworker Info-Sharing‘ .00
LMX -.10* .47** .43“
Inquiry Coworker l Tenure -.25**
Functional Job Status .18" .11** .11**
2 Supervisor Info-Sharing‘ .02
Coworker Info-Sharing‘ .65" .51** .40**
Observation 1 Job Status .19** .03** .03**
Functional
2 Supervisor Info-Sharing‘ -.06
Coworker Info-Sharing‘ .60“ .37** .34**
Experimentation 1 Job Status -.01 .00 .00
Functional
2 Supervisor lnfo-Sharing‘ -.18**
Coworker Info-Sharing‘ .04 .O3* .O3*

131

Table 28 (cont’d.)

 

 

 

 

Dependent Variable Step Predictor B R2 AR2
Inquiry Supervisor 1 Job Status .12 .01 .01
Evaluative
2 Supervisor Info-Sharing” .42“
Coworker Info-Sharing” .27“
LMX -.l6* .30** .29**
Inquiry Coworker 1 Job Status .21** .04“ .04**
Evaluative
2 Supervisor Info-Sharing" -.04
Coworker Info-Sharingb .71** .51** .47**
Experiment 1 Job Status .01 .00 .00
Evaluative
2 Supervisor Info-Sharing" .05
Coworker Info-Sharing” .23“l .06“ .06“

 

*pS.05;**pS.01
‘ Functional information

" Evaluative information

132
Hypothesis 45. Hypothesis 45 examined the relationship between supervisor

and coworker-information sharing and inquiry of coworkers for functional information.
The results of the regression analysis are presented in Table 28. The first step in the
regression equation accounted for a significant amount of variance in the dependent
variable. Tenure was negatively related (B = -.25) and job status was positively
related (B = .18) to inquiry of coworkers for functional information. The second step
of the equation was also statistically significant. Coworker information-sharing was
found to be positively related to inquiry of coworkers for functional information

(B = .65). These results provide partial support for hypothesis 45, in that only
coworker information-sharing was found to be related to the dependent variable.

Hymthesis 46. Hypothesis 46 examined the relationship between supervisor
and coworker information-sharing and observation for functional information. The
regression analysis results are presented in Table 28. The first step in the equation,
the job status covariate, was statistically significant (B = .19). In addition, the second
step in the equation also reached statistical significance. Coworker information-
sharing was the important contributor to this step in the equation (B = .60). These
results provide partial support for hypothesis 46, in that only coworker information-
sharing was found to be related to observation for functional information.

Hyppthesis 47 . Hypothesis 47 examined the relationship between supervisor
sharing and coworker sharing of functional information, and experimentation for
functional information. Regression results are displayed in Table 28. The first step in
the equation, the job status covariate, did not reach statistical significance. The second

step in the equation did account for a significant amount of variance in the dependent

133

variable. Supervisor information-sharing was negatively related to experimentation for
functional information (B = -.l8). This result was opposite to what was predicted by
hypothesis 47.

Hymtheses 48 and 53. Hypotheses 48 and 53 examined the relationship
between supervisor sharing and coworker sharing of evaluative information, LMX, and
inquiry of supervisors for evaluative information. Regression results are presented in
Table 28. The first step in the equation, the job status covariate, did not reach
statistical significance. The second step in the equation did reach statistical
significance. Both supervisor information-sharing (B = .42) and coworker information-
sharing (B = .27) were positively related to inquiry of supervisors for evaluative
information. On the other hand, LMX was negatively related to the dependent
variable (B = -.16). Similar to the results for inquiry of supervisors for functional
information, LMX acted as a suppressor variable in these analyses as well. The
results of this regression support hypothesis 48. Hypothesis 53 was not supported.

Hypothesis 49. Hypothesis 49 examined how supervisor sharing and coworker
sharing of evaluative information were related to inquiry of coworkers for evaluative
information. Table 28 presents the regression results for this hypothesis. Step one,
the job status covariate, was statistically significant (B = .21). The second step in the
equation also reached statistical significance. Coworker information-sharing was
positively related to inquiry of coworkers for evaluative information (B = .71). These
results provide partial support for hypothesis 49, in that only coworker infonnation-

sharing was found to predict the dependent variable.

134
Hymthesis 51. Hypothesis 51 examined the relationship between supervisor

sharing and coworker sharing of evaluative information, and experimentation for
evaluative information. Regression results are presented in Table 28. The first step in
the equation, the job status covariate, was not statistically significant. The second step
in the equation did reach statistical significance. Coworker information-sharing was
positively related to experimentation for evaluative information (B = .23). These
results provide partial support for hypothesis 51.
Information-Seeking and Socialization Outcomes

Hypotheses 56-63 examine relationships between information-seeking strategies
and socialization outcomes. Coworker ratings of mastery, innovation, and workgroup
influence were collected as outcome measures in these analyses. However, newcomers
also provided self ratings of mastery, innovation, and workgroup influence. There was
low consistency between the newcomer and coworker ratings as follows: self and
coworker ratings of mastery were correlated .18; self and coworker ratings of
innovation were correlated .38; and self and coworker ratings of workgroup influence
were correlated .27. It was decided that the analyses for hypotheses 56-63 would be
conducted twice, once using coworker ratings of the outcome variables, and once
using newcomer ratings. This decision was based on the low correlations between
newcomer and coworker ratings of the outcome variables. Because the hypotheses
were based on past research using self ratings, it was not clear whether the same
relationships would be found in the present study where there was low consistency
between self and coworker outcome ratings. Therefore, the present study compared

the relationships between information-seeking strategies and both coworker and

135

newcomer perceptions of the outcome variables. Thus, it could be determined whether
the present study found similar relationships with self ratings as predicted by past
socialization research, and whether these relationships would be found when an
independent source provided the outcome measures.

As in the previous regression analyses, job status was first entered into each of
the analyses as a covariate. In addition, tenure was also entered as a covariate for all
analyses concerning socialization outcomes. It was expected that outcomes such as
mastery, innovation, and workgroup influence would change with increased time on
the job. Therefore, the influence of tenure was controlled before testing the study
hypotheses.

Hymtheses 56-58. Hypotheses 56-58 examined the relationship between
functional information-seeking strategies and mastery of the job. It was predicted that
functional information-seeking across all strategies would be positively related to
mastery; in addition, it was hypothesized that tenure would moderate the effect of two
strategies, inquiry of supervisors and inquiry of coworkers, on mastery. The results of
the moderated hierarchical regression analysis to test these hypotheses are presented in
Table 29, first using self ratings of mastery as the dependent variable, next using
coworker ratings as the dependent variable.

Results using self ratings of mastery indicate that the first step in the equation,
the job status covariates, accounted for a significant amount of variance in the
dependent variable. Tenure (B = .18) was positively related, and job status (B = -.23)
was negatively related to self ratings of mastery. In addition, the second step in the

equation, the linear effects of the information-seeking strategies, also reached statistical

136
Table 29

Functional Information-Seeking and Mastery

A. Mastery - Self Rating

 

 

 

 

 

Step Predictors B R2 AR2
1 Tenure .18**
Job Status -.23** .10** .10**
2 Inquiry of Supervisor - Functional .00
Inquiry of Coworker - Functional -.24**
Observation - Functional .16*
Experimentation - Functional .13* .15** .05**
3 Inquiry of Supervisor x Tenure -.24
Inquiry of Coworker x Tenure .17 .16“ .01
B. Mastery - Coworker Rating
Step Predictors B R2 AR2
1 Tenure .1 1
Job Status -.09 .02 .02
2 Inquiry of Supervisor - Functional -.O2
Inquiry of Coworker - Functional -.l8*
Observation - Functional .05
Experimentation - Functional .06 .05* .03
3 Inquiry of Supervisor x Tenure .73"
Inquiry of Coworker x Tenure -.26 .09** .04**

 

*pS.05;**pS.Ol

137

significance. Experimentation (B = .13) and observation (B = .16) for functional
information were positively related to mastery, while inquiry of coworkers for
functional information (B = -.24) was negatively related to mastery.

The third step in the regression equation, the interaction between tenure and
inquiry of supervisors, and tenure and inquiry of coworkers, did not reach statistical
significance. Hypothesis 56 had mixed support from these results, while hypotheses
57 and 58 were not supported by the results.

A different set of results was found when coworkers provided ratings of
mastery. The first step in the equation did not reach statistical significance. The
second step in the equation, the four information-seeking strategies, also failed to
reach statistical significance. However, the third step in the regression, the interaction
terms, did reach statistical significance. It was found that inquiry of supervisors and
tenure had an interactive effect on mastery (B = .73). Figure 3 displays the
interaction. This figure indicates that, for individuals with little tenure on the job,
inquiry of supervisors for functional information is negatively related to mastery. For
those with greater tenure, inquiry of supervisors for functional information is
positively related to task mastery. These results are directly opposite to those
predicted by hypothesis 57. In addition, the results of this regression analysis did not
support hypotheses 56 and 58.

Hymthesis 59. Hypothesis 59 predicted a positive relationship between all
strategies for seeking evaluative information and mastery of the job. The results of
the regression analyses testing this hypothesis are presented in Table 30. For analyses

using self ratings as the dependent variable, the covariate block was statistically

138

 

0|
1
r

 

—'— Tenure 1

—-k— Tenure 3

 

 

 

Coworker Rating of Mastery
(a) .5

N
r
I

 

1 i I
1 3 5

Inquiry of Supervisor -
Functional information

 

Figure 3. Inquiry of Supervisor (Functional) xTenure Interaction

139
Table 30

Evaluative Information-Seeking and Mastery

A. Mastery - Self Rating

 

 

Step Predictors B R2 AR2
1 Tenure .18**
Job Status -.23** .10“ .10**
2 Inquiry of Supervisor - Evaluative .03
Inquiry of Coworker - Evaluative .09
Experimentation - Evaluative .25“ .18” .08"

 

B. Mastery - Coworker Rating

 

 

Step Predictors B R2 AR2
1 Tenure .11
Job Status -.09 .02 .02
2 Inquiry of Supervisor - Evaluative .00
Inquiry of Coworker - Evaluative .06
Experimentation - Evaluative -.02 .03* .00‘

 

*pS.05;**pS.Ol

' The AR2 values do not add up to the total R2 due to rounding.

140

significant, as in the regression analysis for functional information-seeking. In
addition, the second step in the equation, the three information-seeking strategies, also
reached statistical significance. Experimentation for evaluative information (B = .25)
was positively related to mastery of the job. These results provide partial support for
hypothesis 59.

When coworker ratings of mastery were used, no significant results for the
covariates or information-seeking strategies emerged. Both the first and second steps
in the regression equation failed to reach statistical significance. These results do not
support hypothesis 59.

Hypothesis 60. Hypothesis 60 predicted that experimentation and observation
would account for unique variance in innovation above and beyond inquiry of
supervisors and coworkers, while inquiry of supervisors and coworkers would not
account for unique variance in innovation beyond experimentation and observation.
Two regression analyses were run, reversing the entry of the two sets of information-
seeking strategies, to test these hypotheses. Table 31 presents the regression results.

When self ratings of innovation were the dependent variable, the first step in
the equation reached statistical significance. Tenure (B = .28) was positively related to
innovation. The second step in the equation, inquiry of supervisors and coworkers for
functional information, did not reach statistical significance. The third step in the
equation, experimentation and observation for functional information, did account for a
significant amount of variance in innovation. Experimentation for functional

information (B = .23) was positively related to innovation.

141

When the order of entry of the two sets of information-seeking strategies was
reversed, observation and experimentation become the second step in the regression
equation. This step did reach statistical significance, and experimentation (B = .24)
was positively related to innovation. In addition, the third step in the equation, inquiry
of supervisors and coworkers for functional information, also reached statistical
significance. Inquiry of coworkers for functional information (B = -.18) was
negatively related to innovation.

The failure for inquiry of coworkers to reach significance when entered first
but to become significant when entered second is due to the presence of observation as
a suppressor variable. Observation and inquiry of coworkers for functional
information were highly intercorrelated. Observation was almost uncorrelated with
innovation (r = -.01). In addition, when it was entered first in the regression, its
beta-weight was small and positive (B = .02). However, when inquiry of coworkers
was entered, the beta—weight for observation remained positive and became stronger
(B = .12). Therefore, it may have acted as a suppressor variable by removing some of
the variance in inquiry of coworkers that was not associated with the dependent
variable, allowing the relationship to become significant when entered last in the
regression.

These results provide partial support for hypothesis 60. Experimentation for
functional information did account for unique variance in innovation above and
beyond that accounted for by inquiry of supervisors and coworkers. On the other
hand, it was also found that inquiry of coworkers for functional information accounted

for unique variance in innovation above and beyond that accounted for by

142
Table 31

Functional Information-Seeking and Innovation

A. Innovation - Self Rating

 

 

Step Predictors B R2 AR2
1 Tenure .28"
Job Status .07 .08" .08“
—
2 Inquiry of Supervisor - Functional -.07
Inquiry of Coworker - Functional -.10 .10** .02‘
3 Observation - Functional .12
Experimentation - Functional .23“ .16" .06"
—
2 Observation - Functional .02
Experimentation - Functional .24“ .14“ .O6**
3 Inquiry of Supervisor - Functional -.04
Inquiry of Coworker - Functional -.18"‘ .16“ 02*"

 

B. Innovation - Coworker Rating

 

 

 

 

Step Predictors B R2 AR2

1 Tenure .24“

Job Status .13* .06" .06"
2 Inquiry of Supervisor - Functional -.01

Inquiry of Coworker - Functional -.12 .08** .01
3 Observation - Functional .01

Experimentation - Functional .14* .10” .02
2 Observation - Functional -.O7

Experimentation - Functional .15* .09“ .03*
3 Inquiry of Supervisor - Functional .01

Inquiry of Coworker - Functional -.13 .10" .01

 

“ p S .05; ** p S .01.
' exact value = .018.
” exact value = .023.

143

experimentation and observation. However, it was negatively related to innovation, which
is consistent with the notion that individuals who rely on their own experimentation and
less on other workgroup members will tend to be more innovative.

Table 31 also presents the hierarchical regression results when coworker ratings of
innovation were used as the dependent variable. The first step of the equation, the
covariates, did reach statistical significance. Both tenure (B = .24) and job status
(B = .13) were positively related to innovation. The second step in the equation, inquiry
of supervisors and coworkers for functional information, did not reach statistical
significance. The third step in the equation, experimentation and observation for
functional information, also failed to reach statistical significance.

When the order of entry of the two information-seeking strategy blocks was
reversed, the second step in the equation did reach statistical significance.
Experimentation for functional information (B = .15) was positively related to coworker
ratings of innovation. The third step in the equation, inquiry of supervisors and coworkers
for functional information, did not reach statistical significance. Thus, the results for the
coworker ratings of innovation were more ambiguous than those using self ratings.
Specifically, experimentation and observation did not account for unique variance in
innovation above and beyond inquiry of supervisors and coworkers. However,
experimentation did account for a significant amount of total variance in innovation and
was related to innovation in the predicted, positive direction.

Hypothesis 61. Hypothesis 61 predicted that seeking evaluative information across
all strategies would be positively related to innovation. Table 32 presents the regression
results testing this hypothesis. Using newcomer ratings of innovation, the first step in the

equation, the 'covariates, was statistically significant (as in the results for hypothesis 60).

144
Table 32

Evaluative Information-Seeking and Innovation

A. Innovation - Self Rating

 

 

 

 

 

Step Predictors B R2 AR2
1 Tenure .28“
Job Status .07 .08“ .08“
2 Inquiry of Supervisor - Evaluative .05
Inquiry of Coworker - Evaluative .04
Experimentation - Evaluative .14* .11** .03*
B. Innovation - Coworker Rating
Step Predictors B R2 AR2
1 Tenure .24“
Job Status .13* .06“ .06”
2 Inquiry of Supervisor - Evaluative .05
Inquiry of Coworker - Evaluative .03
Experimentation - Evaluative .04 .07“ .01

 

*pS.05;**pS.01

145

In addition, the second step in the equation, the information-seeking strategies,
also accounted for a significant amount of variance in innovation. Experimentation for
evaluative information (B = .14) was positively related to innovation. These results
provide partial support for hypothesis 61.

When coworker ratings of innovation were used, the covariates accounted for a
significant amount of variance in the dependent variable (as in the results for
hypothesis 60). However, the second step in the equation did not reach statistical
significance. The information-seeking strategies for evaluative information were not
related to coworker ratings of innovation, providing no support for hypothesis 61.

Hypotheses 62 and 63. Hypotheses 62 and 63 predicted that mastery would be
positively related to workgroup influence, and that mastery would moderate the
relationship between innovation and workgroup influence, respectively. Table 33
presents the moderated hierarchical regression results testing these hypotheses.

When self ratings of the three outcome variables were used, the first step in the
equation did not reach statistical significance. The second step in the equation, the
linear effects of mastery and innovation, did reach statistical significance.

Specifically, both mastery (B = .37) and innovation (B = .15) were positively related to
workgroup influence. The third step in the equation, testing the interactive effects of
mastery and innovation, did not reach statistical significance. These results provided
support for hypothesis 62. Hypothesis 63 was not supported, although innovation was

related to workgroup influence in the predicted direction.

146

When coworker ratings of the three outcome measures were used, similar
results were obtained. The first step in the equation, the covariates, did reach
statistical significance. Both tenure (B = .23) and job status (B = .18) were positively
related to workgroup influence. In addition, the second step in the equation, the linear
effects of mastery and innovation, did account for a statistically significant amount of
variance in workgroup influence. Both mastery (B = .34) and innovation (B = .24)
were positively related to workgroup influence. The third step in the equation, the
interaction between mastery and innovation, did not reach statistical significance.
Thus, like the analysis with self ratings, hypothesis 62 was supported and hypothesis

63 was not supported by the regression results.

147
Table 33

Mastery, Innovation, and Workgroup Influence

A. Self Ratings

 

 

 

 

 

Step Predictors B R2 AR2
1 Tenure .10
Job Status .09 .01 .01
2 Mastery .37**
Innovation .15* .17** .16**
3 Mastery x Innovation -.09 .17“ .00
B. Coworker Ratings
Step Predictors B R2 AR2
1 Tenure .23"
Job Status .18“ .07** .07**
2 Mastery .34“
Innovation .24M .28** .21“
3 Mastery x Innovation .45 .28“ .00

 

*ps.05;**ps.01

DISCUSSION

The purpose of this study was to examine how new employees are active in
seeking functional and evaluative information in order to adapt to their new jobs. In
addition, the research was conducted to identify antecedent influences of the
information-seeking process, as well as demonstrate how information-seeking is related
to socialization outcomes.
Patterns of Information-Seeking

The present study was able to shed some light on the process of newcomer
adaptation. First, the research examined the relative frequency with which new
employees use different information-seeking strategies. Results indicated that
individuals rely differentially on information-seeking strategies for functional and
evaluative information. Experimentation was the most frequently used strategy for
both types of information. For functional information, coworkers were relied on more
frequently than supervisors for those with very little time on the job. However, with
increased tenure on the job, this difference between the fi'equency of inquiry of
supervisors and inquiry of coworkers disappeared. These shifts in the use of strategies
over time are similar to those observed in the Ostroff & Kozlowski (1992) longitudinal
socialization study.

For evaluative information, supervisors were asked for information more
frequently than coworkers. However, for both functional and evaluative information,

the difference between inquiry of supervisors and coworkers was very small. The

148

149

more important difference seems to be that new employees rely the most on their own

experimentation to learn how to perform their jobs. This certainly makes sense from a
training perspective in that practice or hands-on experience is critical to learning (e.g.,

Baldwin & Ford, 1988; Kozlowski, Ford, & Smith, 1993).

In addition, the results indicated that inquiry of supervisors and coworkers for
functional information decreased with increased time on the job. As newcomers begin
to learn the basics of their job, they rely less on acquiring information from other
organizational members. On the other hand, experimentation remains at its relatively
high level, indicating that new employees are still trying out methods to perform their
job on their own. Thus, they have become more self-sufficient in handling the
demands and difficulties of their new position.

The study did not find that inquiry of supervisors and coworkers for evaluative
information increased over time. Instead, it remained at a constant rate. The social
costs of asking for evaluative information may prevent employees from asking for
evaluative information too frequently (Ashford & Cummings, 1983). On the other
hand, the present study provides a cross-sectional description of the learning process;
research that examines within—person changes over time may show different results.
Antecedent Factors and Information-Seeking

This study also examined factors that may influence newcomer information—
seeking. First, seven regression analyses examined how job characteristics and self-
efficacy influenced the use of information-seeking strategies. The four infonnation-
seeking strategies were conceptualized as falling roughly into two groups: inquiry of

supervisors and inquiry of coworkers as one group, and observation and

150

experimentation as the other group. Observation and experimentation were grouped
together from a social learning perspective, and were thought to be more self-sufficient
strategies than inquiry. Hypotheses suggested that certain job characteristics would
made it more likely for individuals to use inquiry of supervisors and coworkers for
information. In contrast, it was hypothesized that other job characteristics made it
more likely for individuals to use observation and experimentation.

However, the intercorrelations between the information-seeking strategies
indicated that observation was more closely related to inquiry of coworkers. It appears
that, when individuals have Opportunities to ask coworkers questions, they also have
greater opportunities to observe the behavior of others. Thus, as the high
intercorrelation between observation and inquiry of coworkers suggested, job
characteristics tended to relate similarly to these two information-seeking strategies.

Results of the regression analyses found that certain job characteristics did
influence information-seeking. In addition, self—efficacy did impact information-
seeking where it could be expected to have its strongest influence. Specifically,
interdependence was found to positively influence the following information-seeking
strategies: inquiry of coworkers for functional information; observation for functional
information; inquiry of supervisors for evaluative information; inquiry of coworkers
for evaluative information; and experimentation for evaluative information.

The results for inquiry of supervisors and coworkers were in the direction
predicted by the study hypotheses. When new employees had to interact with other
employees to get work accomplished, they had more opportunities to ask questions and

clear up any difficulties. In addition, there may have been social pressures due to

151
interdependence that urged new employees to take advantage of the opportunity to ask

questions. The positive relationship found with observation for functional information
is also consistent with this explanation. Individuals who interacted more frequently
with others due to the nature of their jobs had more opportunities to observe their
behavior. 1

On the other hand, interdependence was positively related to experimentation
for evaluative information. It was predicted that individuals in jobs with low
interdependence would have to rely more on evaluating their own performance, since
they would have less opportunities to interact with others and ask for their feedback.
Instead, greater interdependence led to greater self-evaluation. It may be that
individuals felt more responsible for their work because other workgroup members
relied on them, and thus were more proactive in evaluating themselves and insuring
that they were doing a good job.

Discretion was also an important job characteristic variable that affected
information— seeln'ng. Discretion was negatively related to inquiry of supervisors for
functional information, and positively related to experimentation for both functional
and evaluative information. These relationships were predicted by the study
hypotheses. When new employees entered jobs in which they had the latitude to
determine how to accomplish work, they had to rely on their own efforts to learn the
job and evaluate their work. Asking supervisors for functional information might
show that they could not handle the responsibility of developing their own work
methods. In addition, other sources might be unable to provide them with the required

information for a job with no standard procedures. Because individuals in high

152

discretion jobs must rely on experimentation to learn their tasks, they are more likely
to engage in role innovation.

One aspect of job complexity, exceptions, influenced the use of three
information-seeking strategies. Exceptions was positively related to observation for
functional information, and to inquiry of both supervisors and coworkers for evaluative
information. The results for inquiry of supervisors and coworkers are consistent with
a learning perspective. When new employees are presented with a wide variety of
tasks they must perform as part of their job, then it is more difficult for them to
determine how well they are performing. Therefore, asking their supervisor and
coworkers for evaluative information may help them to understand where their job
performance needs improvement. Although the positive relationship between
exceptions and observation for functional information was opposite to that predicted, it
is still understandable from a learning perspective. The greater the variety in the types
of tasks performed, the more difficult it may be to learn the job without assistance.
Watching how others handle this variety is an effective way to adapt to the job.

Analyzability was related to only one information-seeking strategy.
Analyzability was positively related to inquiry of supervisors for evaluative
information. This relationship was opposite to that predicted. The hypothesis
predicted that, when jobs were less analyzable, individuals would be less able to
understand their work and would have to rely more on the input of others to
understand how well they were performing. When the job was more analyzable,
individuals would understand their job duties better, and would not have to ask others

for evaluative information. However, the results indicated that individuals asked

153

supervisors for evaluative information more frequently when jobs were more
analyzable. This result is hard to explain fi'om a learning perspective. It may be the
result of the suppressor effect described in the results section. It may also be
understandable from an impression management point of view (Morrison & Bies,
1991). In more complex jobs, new employees are likely to be making mistakes, and
they may not want to call attention to them by asking for evaluative information. On
the other hand, in less complex jobs, individuals may learn their tasks relatively
quickly. They may then ask their supervisor for evaluative information more
frequently to attract attention to their good performance.

Finally, self-efficacy was found to have main effects on several of the
information-seeking strategies. Self-efficacy was negatively related to inquiry of
supervisors and inquiry of coworkers for functional information, as predicted by the
hypotheses. These results are consistent with previous research (Major & Kozlowski,
1990). In addition, self-efficacy was positively related to experimentation for
evaluative information, in a direction opposite to that hypothesized.

These results indicate that, for functional information, individuals who are less
confident of their ability to succeed on the job do rely more on other individuals to
learn how to perform their tasks. Because they feel they do not have the expertise to
succeed at their tasks, they rely on the expertise of others to show them the best way
to get their work accomplished. In contrast, new employees who are more confident
of their abilities rely less on other individuals to acquire functional information. They
may have previous experiences similar to their new job that increase their confidence

and provide them with methods to perform their jobs.

154
It was hypothesized that individuals low on self-efficacy would be less likely to

ask others for information because of the social costs of appearing incompetent.
However, it was assumed that evaluative information would still be perceived as
important by newcomers, so that they would find some means to acquire it. Thus, it
was expected that they would acquire this information in a less costly manner, through
self-evaluation. The results seem to indicate that self-evaluation is also costly, because
low self-efficacy individuals were less willing to evaluate their own performance
compared to high self-efficacy individuals. Individuals who feel little confidence in
their performance may avoid self-evaluations, because they know they are not doing as
well as they would like. This may be problematic from the position that infonnation-
seeking leads to greater learning and mastery of the job. This failure to seek
evaluative information may limit the potential for low self-efficacy individuals to
improve their performance, unless supervisors and coworkers are proactive in
providing evaluative information to them.

Unfortunately, the regression analyses failed to find the hypothesized
interaction effects between job characteristics and self-efficacy. A major problem with
the analyses was the lack of statistical power to detect these moderator relationships.
A power analysis estimated that a sample size of 348 would be necessary for an 80
percent probability that a block of interactions, accounting for 3 percent of the
variance, would be statistically significant (p < .05) beyond the estimated 15 percent
of the variance accounted for by the main effects of job characteristics and self-
efficacy. This study attempted to obtain a sample of this size; however, the actual

sample obtained was almost one-third smaller than that required by the power analysis.

155
In addition, the analyses failed to support many of the hypotheses specified.

Statistical power was one problem. Furthermore, the information-seeking strategies
were not compensatory as expected (i.e, experimentation did not compensate for the
lack of opportunities for inquiry). Therefore, many of the opposing relationships for
inquiry v. experimentation were not found. On the other hand, job characteristics and
self-efficacy did influence the information-seeking strategies where they could be
expected to have the most impact. Interdependence was found to mainly influence
strategies involving interactions with others. Discretion had its major influence on
experimentation. In addition, self-efficacy did have an influence on inquiry of
supervisors and inquiry of coworkers, where the costs of appearing incompetent could
be expected to be higher.

This research also examined how interpersonal relationships between new
employees, their supervisor, and coworkers would impact their information-seeking.
In general, results indicated that, when supervisors and coworkers were more proactive
in providing functional and evaluative information, newcomers were more willing to
actively seek this information on their own. This is consistent with research indicating
that workgroups have norms for information-seeking (e.g., Dewhirst, 1971). In
addition, the relationships between coworker and supervisor information-sharing and
newcomer information-seeking tended to be specific to the strategy involved. For
example, supervisor information-sharing was critical for inquiry of supervisors for
functional information. Coworker information-sharing had a positive impact on
inquiry of coworkers for both functional and evaluative information, and was the

important contributor to observation for functional information as well.

156

On the other hand, both supervisor and coworker information-sharing positively
influenced inquiry of supervisors for evaluative information. For experimentation,
supervisor information-sharing was negatively related to experimentation for functional
information, while coworker-information sharing was positively related to
experimentation for evaluative information.

The study hypotheses were more general in their predictions, specifying that
both supervisor and coworker information-sharing would be positively related to the
information-seeking strategies. However, it appears that supervisor and coworker
willingness to share information influence different information-seeking strategies. It
is easy to explain that, for example, supervisor information-sharing should be the
important influence on inquiry of supervisors. It is less easy to explain why
supervisor v. coworker information-sharing has the impact on experimentation, for
functional v. evaluative information, respectively.

Nonetheless, when newcomers perceive that other workgroup members are
willing to share information, they are more proactive in asking for information they
need. Thus, newcomers do not seek information more frequently when others fail to
provide information, which might be expected from the perspective that information is
a valued and scarce resource (Ashford & Taylor, 1990; Jablin, 1984). It appears
instead that supervisor and coworker willingness to provide information operate more
like a norm or climate for information-sharing. The one exception to this finding is
that newcomers are less likely to experiment on their own when their supervisors
provide them with information on how to perform their jobs. This result suggests that,

if supervisors are very directive about how to accomplish work, newcomers will be

157
less likely to experiment and find their own methods for performing work. Thus, if

supervisors are too explicit in how they want work accomplished, newcomers are less
likely to be innovative on the job.

This study also hypothesized that greater leader-member exchange (LMX)
relationships would lead to more fiequent inquiry of supervisors for functional and
evaluative information. However, the results indicated that the quality of the
relationship that developed between new employees and their supervisors had little
impact on information-seeking. LMX was not significantly correlated with inquiry of
supervisors for functional or evaluative information, and it acted as a suppressor
variable in the regression analyses. It appears that supervisor information-sharing is
the important influence on the willingness of newcomers to ask their supervisor for
information.

For the analyses examining how job characteristics, self-efficacy, and
workgroup relationship variables affected information-seeking, the results were based
on percept-percept relationships. Thus, common method bias is a concern. However,
intercorrelations between job characteristics and information-seeking strategies ranged
from -.26 to .39. Intercorrelations between self-efficacy and information-seeking
strategies ranged fiom —.27 to .29. Finally, LMX was correlated with information-
seeking strategies from -.06 to .12. These intercorrelations do not appear too extreme,
and the higher intercorrelations do tend to be found where hypothesized or where they
make the most conceptual sense. Therefore, for these variables, method variance may

be less of a concern.

158

On the other hand, as mentioned in the results, supervisor and coworker
information-sharing were highly correlated with the information-seeking strategies Of
inquiry of supervisors and coworkers. The content of these variables was the same,
functional and evaluative information, which may indicate a method bias problem.
However, the survey instructions were explicit in indicating whether it was the
newcomer who was active in asking for the information, or it was the supervisor or
coworker who actively approached the new employee and provided the information.
In addition, while the large relationships found between supervisor and coworker
information-sharing and inquiry of supervisors and coworkers may be partly
attributable to method variance, they are still consistent with recent research indicating
similar results (Major, 1990; Morrison, 1992).

Information-Seeking and Socialization Outcomes

Analyses examining how information-seeking related to socialization outcomes
was able to deal with common method bias by examining how information-seeking
relates to coworker ratings of socialization outcomes. However, because past research
has used self ratings of outcome measures, the analyses were run with both self and
coworker ratings to determine if different results would emerge.

When self ratings of mastery were used, experimentation and observation for
functional information were positively related to mastery, as predicted by the
hypotheses. This is consistent with previous socialization research (Ostroff &
Kozlowski, 1992). In addition, inquiry of coworkers for functional information was

negatively related to mastery. This was Opposite to the relationship predicted, but

159

consistent with Morrison’s (1993) findings. In addition, the interactions between
inquiry of supervisors and coworkers, and job tenure, were not significant.

Individuals who were more self-sufficient in learning their jobs through their
own observation and experimentation exhibited greater mastery of the job. On the
other hand, individuals who asked their coworkers for functional information more
frequently exhibited lower mastery. These results are cross-sectional, and it may be
that individuals who are not doing very well need to seek more assistance from their
coworkers. It is unclear whether this same relationship would hold if mastery was
measured at a later point in time.

When coworker ratings of mastery were the dependent variable, there were no
significant main effects for information-seeking. Inquiry Of supervisors did interact
with job tenure to predict mastery, but in a direction Opposite to that predicted. For
individuals low on job tenure, inquiry of supervisors for functional information was
negatively related to mastery. The relationship was positive for individuals with
greater job tenure. Coworkers may perceive that individuals who rely to a greater
extent on information from their supervisor do not know their job as well at the
present time. However, if one were to measure mastery at a later time, these
individuals may be rated as having greater mastery of their job. The information they
have acquired at the present time may ultimately lead to greater mastery when
measured several months later.

When self ratings were used, results indicated that experimentation for
evaluative information was positively related to mastery of the job. Individuals who

were more active in evaluating their own performance were also more successful at

160

performing their jobs. However, the results using coworker ratings of mastery
indicated no significant relationships between information-seeking for evaluative
information and mastery.

When examining how information-seeking for functional information was
related to innovation, some significant results were found using both self and coworker
ratings as dependent variables. It was hypothesized that experimentation and
observation for functional information would account for unique variance in innovation
above and beyond the influence of inquiry of supervisors and coworkers on
innovation. When self ratings of innovation were used, experimentation for functional
information accounted for unique variance in innovation. However, when the order of
entry of the information-seeking strategies was reversed, inquiry of coworkers for
functional information was also found to account for unique variance in innovation.

As explained in the results section, inquiry of coworkers became significant when
entered last in the regression because observation acted as a suppressor variable.

Experimentation for functional information was positively related to innovation,
and inquiry of coworkers was negatively related to innovation. While these results did
not fully support the hypotheses, they are consistent with the conceptual ideas
developed in this paper. As expected, individuals who more actively experiment to
learn their job find more innovative ways of performing their tasks. On the other
hand, individuals who rely to a greater extent on coworkers to discover how to do
their work are less likely to perform their tasks in new or innovative ways.

When coworker ratings of innovation were the dependent variable, neither

group of information-seeking strategies accounted for unique variance beyond the

161

other group of strategies. However, experimentation for functional information did
account for a significant amount of total variance in the dependent variable.
Experimentation for functional information was found to be positively related to
coworker ratings Of innovation.

Analyses examining the relationship between evaluative information-seeking
and self ratings of innovation indicated that experimentation for evaluative information
was positively related to innovation. Individuals who monitored their own
performance more frequently were more innovative on the job. However, when
coworker ratings of innovation were the dependent variable, no information-seeking
strategies were found to significantly predict innovation.

Finally, when using both self and coworker ratings of the outcome variables,
mastery and innovation were positively related to workgroup influence. However, the
hypothesized interaction between mastery and innovation was not significant. Each
factor had a positive main effect on workgroup influence. Individuals who had
learned their jobs, and had found better ways to get their work accomplished, were
able to participate in decision-making and provide suggestions to their new workgroup.

An obvious aspect of these results is that, when using self ratings of outcome
measures, a number of the hypothesized relationships were found, and were consistent
with previous literature on newcomer socialization and adaptation. On the other hand,
few significant results were found when using coworker ratings of outcome measures.
This may be partly attributable to the low reliabilities obtained when averaging the

responses from two or three coworkers. This low consistency between coworkers

162

when rating the same newcomer decreased the power to detect any true relationships
between information-seeking and socialization outcomes.

These findings also highlight the difficulty of determining who is in the best
position to provide outcome ratings. For example, one may be interested in newcomer
perceptions of outcome measures in order to identify those information-seeking
strategies that lead new employees to consider that they have mastered their jobs to a
higher degree. This would be important from the position that enhancing mastery, for
example, will lead to increased self-efficacy. Several studies found that self-efficacy
leads to higher performance in learning settings (e.g., Gist, 1987; Gist et al., 1989).

In contrast, the concern for method variance may require an independent source
for acquiring outcome measures. Although newcomers report that coworkers are
available most of the time (e.g., Louis, et al., 1983), they may not be the best source
for these ratings. For example, coworkers who have been on the job longer than
newcomers may use a different frame of reference when rating new employees. Even
when using the same rating scale, these individuals may have different standards for
excellence. In addition, supervisors have traditionally been used as the appropriate
source for outcome ratings. Therefore, future research should use supervisor ratings,
as well as determine whether supervisors and coworkers tend to exhibit greater
consistency and agreement in their ratings than coworkers and newcomers.

Furthermore, the traditional socialization outcomes such as mastery may not
necessarily be the most appropriate criteria when examining the learning process.
Feldman (1989) has noted that the training and socialization literatures would benefit

from integrating some of their ideas. The training literature may suggest some

163

alternate outcome measures to the traditional mastery criterion. For example, in the
training literature, traditional behavioral measures have been joined by more recent
criteria that measure cognitive outcomes such as: (l) the individual’s mental models,
or how their knowledge is structured and organized, and (2) metacognition, or the
knowledge and regulation of one’s cognitions (Kraiger, Ford, & Salas, 1993). The
training literature suggests that experts and novices have different mental models of
the job; this may be one explanation for the lack of consistency between newcomer
and coworker ratings of outcome measures. While newcomers may have a relatively
simple view of their tasks, coworkers are likely to have a much more detailed mental
model of what is required on the job.

Further, when measuring skill-based outcomes, the training literature recognizes
that there are stages of skill acquisition: (1) initial skill acquisition; (2) skill
compilation; and (3) skill automaticity. These stages of skill acquisition can be
assessed through observation, hands-on tests, and structured situational interviews
(Kraiger et al., 1993). For researchers interested in the learning process during early
socialization, these types of outcome measures may be more sensitive and meaningful
than more general ratings scales evaluating mastery.

Nonetheless, research on newcomer socialization is difficult because there is a
trade-off between capturing the learning process at a level that is generalizable, and
understanding the specific learning process that occurs for individual jobs. For
example, a number of early socialization studies have focused on only one job at a
time (Feldman, 1989), and thus were able to use more sensitive outcome measures

such as interviews and observation that were specific to the job in question. However,

164

if one is interested in understanding the adaptation process across a wide variety of
jobs, then it is difficult to develop criteria that are sensitive to the specific challenges
and difficulties of each job.

Sr_r_m_m__ary and Implications

The present study has helped to explain the process through which individuals
learn and acquire information early on the job. Self-efficacy, and the job
characteristics of interdependence, discretion, and exceptions, influence the strategies
used by newcomers to acquire information. In addition, supervisor and coworker
information-sharing were important in increasing newcomer proaction in acquiring
information. Thus, supervisors and coworkers are powerful socialization agents not
only in providing critical information, but also in establishing a climate where new
employees feel it is acceptable to ask questions when there is something they do not
understand. Organizations may want to sensitize their members to the particular needs
and concerns of new employees. This will not only produce a more positive learning
atmosphere, but may also facilitate the newcomer’s job proficiency and assimilation
into the workgroup.

This study also indicates that information-seeking is related to newcomer
perceptions of mastery and innovation. Particularly important are newcomer attempts
to experiment on their own and monitor and evaluate their own performance.
Organizations may want to encourage new employees to experiment with methods to
accomplish their tasks. If new employees are explicitly shown only one correct
method for performing their jobs, they may experience less Opportunities to: (1) make

errors, and thus better understand their tasks, and (2) innovate and find better ways to

165

get their work done. The training literature suggests that practice that leads to
conflicts or errors is important for triggering new learning in tasks that are error—prone,
when the tasks are being performed incorrectly, and when the tasks require the
individual to monitor and correct his/her performance (Kozlowski et al., 1993). Of
course, this recommendation must be qualified by the type of job the new employee
acquires; if the individual is in a position that is potentially dangerous or costly, then
explicit and standardized work procedures would of course be required. On the other
hand, in professional jobs where there can be different ways for work to be
accomplished, new employees may benefit fi'om some latitude in experimenting with
their tasks. In fact, socialization researchers often suggest that innovation within an
acceptable range is an indicator of effective socialization for both the individual and
the organization (Feldman, 1989; Schein, 1968).
Directions for Future Research

While this research was successful in identifying some of the antecedents and
outcomes of newcomer information-seeking, there are a number of areas where
additional research is needed. One limitation of the present study is that it was cross-
sectional, providing only a snapshot of the adaptation process. Longitudinal research
is needed to examine how information-seeking changes over time. In addition,
longitudinal research can also examine how information-seeking at an earlier point in
time influences socialization outcomes measured at a later time. For example, while
coworker ratings of outcome measures were not influenced much by information-

seeking, it is possible that the influence may be detected over time.

166

In addition, several interesting moderator relationships were hypothesized
between job characteristics and self-efficacy in predicting information-seeking. While
the present study failed to find groups of these interactions to be significant (as
evidenced by a nonsignificant ARZ), several interactions did result in significant
beta-weights. For example, the beta-weight for the interaction between
interdependence and self-efficacy was significant in predicting inquiry of supervisors
for functional information. This interaction suggests that for individuals high on self-
efficacy, the hypothesized positive relationship between interdependence and inquiry
was found. However, for individuals low on self-efficacy, interdependence was
negatively related to inquiry of supervisors for functional information. It may be that,
for individuals who are not confident about their abilities, the more other workgroup
members rely on them to get their own work done, the less willing they are to let
others know of their inadequacies by asking for needed information.

The beta-weight for the interaction between discretion and self-efficacy in
predicting inquiry of coworkers for functional information was also statistically
significant. For individuals high on self-efficacy, the hypothesized negative
relationship between discretion and inquiry of coworkers for functional information
was found. For individuals low on self-efficacy, the relationship between discretion
and inquiry of coworkers for functional information was positive. It is possible that
newcomers who are low on self-efficacy may find jobs high in discretion to be
stressful; they probably feel little confidence in their ability to develop their own work

procedures and would rather rely on seasoned incumbents who they perceive to have

167

more expertise. Individuals high on self-efficacy may welcome the challenge of a job
high in discretion, and rely less on the help of coworkers in learning their jobs.

Two interactions evidenced significant beta-weights in predicting
experimentation for functional information. First, exceptions and self-efficacy
interacted to predict experimentation. For individuals low on self-efficacy, the
hypothesized negative relationship between exceptions and experimentation for
functional information was found. For individuals high on self-efficacy, exceptions
was positively related to experimentation for functional information. Individuals low
on self-efficacy may find jobs high in variety to be too demanding early on the job;
therefore, they may be less able to figure out on their own how to perform their tasks.
On the other hand, individuals who are confident in their abilities may be able to
handle greater task variety by trying out work methods on their own.

The interaction between analyzability and self-efficacy also showed a
significant beta-weight in predicting experimentation for functional information. For
individuals low on self-efficacy, analyzability was negatively related to
experimentation for functional information, in an opposite direction to that
hypothesized. For individuals high on self-efficacy, there appeared to be no
relationship between analyzability and experimentation for functional information.
Individuals who were highly self-efficacious engaged in relatively high levels of
experimentation for functional information, regardless of the degree of job
analyzability. In contrast, for individuals low on self-efficacy, it was necessary to
engage in high levels of experimentation for functional information when the job was

not easily analyzable.

168

In sum, these interactions suggest some interesting relationships between job
characteristics and information—seeking that change depending on the newcomer’s self-
efficacy. However, the present study had insufficient statistical power to detect these
relationships. In addition, the present study hypothesized interactions between all job
characteristics and self-efficacy, for all information-seeking strategies. Future research
should examine more precise hypotheses that take into account the specific job
characteristic and information-seeking strategy in question.

There are a number of other directions for future research to take. For
example, this study focused on only two types of information that newcomers acquire,
functional and evaluative information related to their jobs. New employees must also
acquire information about relationships with their supervisor and coworkers, and about
the organization’s politics, culture, and values (e.g., Chao, O’Leary, Walz, Klein, &
Gardner, 1990; Ostroff & Kozlowski, 1992). Research should examine the strategies
individuals use to acquire these additional types of information.

Research should also continue to examine individual differences in the use of
information-seeking strategies. The present study indicated that self-efficacy is an
important individual difference factor that impacts information-seeldng. Two
constructs identified as part of the "big five" personality dimensions may also
influence how individuals seek information (e.g., Barrick & Mount, 1991). First,
conscientiousness may lead to greater information-seeking across all strategies.
Second, openness to experience may be an important predictor of an individual’s

willingness to experiment in a new job. In sum, the present study is just an initial

step in understanding how newcomers acquire information, and the factors that

 

169

produce differences in information-seeking. Future research should continue to
identify information that is important to new employees, and the strategies for

acquiring this information that lead to effective adaptation and socialization.

APPENDICES

170
APPENDD( A

Ingm’ of Sumrvisor - Functional Information Items

To what extent do you:

1.

2.

Ask your supervisor how to complete the duties of your job.

Ask your supervisor how to perform your tasks.

Ask your supervisor how to accomplish the different parts of your job.
Ask your supervisor how to establish priorities for your work.‘

Ask your supervisor how to deal with problems on your job.

Ask your supervisor how to use equipment and materials necessary to perform
your job.

Ask your supervisor how to do your job.

Note. ‘Item deleted fi'om the final inquiry of supervisor for functional information
scale.

The rating scale ranged from (1) = gy r_n_f;e_q' uently to (5) = ve_ry figuently.

l7 1
APPENDIX B

Ingm’ of Spmrvisor - Evaluative Information Items

To what extent do you:

1.

2.

Ask your supervisor how well you are doing on the job.

Ask your supervisor how well you have established procedures to get your work
done.

Ask your supervisor how well you are meeting your job responsibilities.

Ask your supervisor how well you are succeeding in meeting your work
objectives.

Ask your supervisor how well you handle difficulties on the job.
Ask your supervisor how well you are performing your tasks and duties.

Ask your supervisor how well you are performing on the job.

Note. The rating scale ranged from (1) = e_ry infrgguently to (5) = veg figuently.

172
APPENDIX C

Ingm’ of Coworkers - Functional Information Items

To what extent do you:

1.

2.

Ask your coworkers how to handle problems on the job.‘

Ask your coworkers how to use equipment and materials necessary to do your
job.‘

Ask your coworkers how to divide your time among different parts of your job.‘
Ask your coworkers how to perform your tasks.

Ask your coworkers how to do your work.

Ask your coworkers how to establish priorities for your work.‘

Ask your coworkers how to complete your job duties.

Note. ‘Items deleted from the final inquiry of coworkers for functional information
scale.

The rating scale ranged from (1) = vepy m uently to (5) = vpg frgguently.

173
APPENDIX D

Ingm' of Coworkers - Evaluativg Information Items

To what extent do you:

1.

Ask your coworkers how well you have established methods for getting your
work done.‘

Ask your coworkers how well you handle your job duties.
Ask your coworkers how well you are performing your tasks.
Ask your coworkers how well you are performing on the job.
Ask your coworkers how well you are doing on the job.

Ask your coworkers how well you are succeeding in meeting work objectives.

Ask your coworkers how well you are meeting the requirements of your job.

Note. 'Item deleted from the final inquiry of coworkers for evaluative information
scale.

The rating scale ranged ham (1) = vgy rn_frg° uently to (5) = vgry figuently.

174
APPENDD( E

Observation - Functional Information Items

 

To what extent do you:

1.

Observe the behavior of others to find out how to handle your job
responsibilities.

Observe those around you to learn how to divide your time among different parts
of your job.

Observe how others work to discover how to perform your tasks.

Observe your coworkers to find out how to use the equipment and materials
necessary to do your work.

Observe others to determine how to establish priorities for your assignments.

Observe your coworkers handling problems to find out how to handle difficulties
on the job.

Note. The rating scale ranged from (1) = vm infiguently to (5) = ve_ry frgguently.

175
APPENDIX F

Observation - Evaluative Informgtion Item_s_

To what extent do you:

1.

Observe your coworkers doing their jobs to find out how well you are performing
your tasks.

Observe how your coworkers act towards you to discover how well they think
you have learned your job.

Determine how well you are meeting the responsibilities of your job by
comparing your performance to that of your coworkers.

Observe the performance of others to compare how well you are doing on the
new job to them.

Observe how your boss rewards others to understand how well he/she would
evaluate your work.

Observe how others behave towards you to determine how well they think you
are performing.

Note: This scale was deleted from the final analyses based on factor analysis results.

The rating scale ranged from (1) = vgg m uently to (5) = vm figuently.

176
APPENDIX G

Em’ ntation - Functional Information Items

To what extent do you:

1.

2.

Experiment on your own to find out how to do your job.
Experiment on your own to discover how to complete your job duties.

Experiment to discover how to solve any problems with tasks or assignments on
your own.

Experiment on your own to learn how to use the equipment and materials
necessary to do your job.

Experiment to find out how to perform your tasks.

Experiment to learn how to accomplish the different parts of your job.

Note: The rating scale ranged from (1) = veg infrgguently to (5) = g fluently.

177
APPENDDI H

Em‘ ntation - Evaluative Information Items

To what extent do you:

1. Evaluate your own work to determine how well you are meeting your work
Objectives.

2. Evaluate how well you are accomplishing the responsibilities of your job.

3. Evaluate your performance of tasks and duties to determine how well you are
doing on the job.

4. Evaluate how well you have learned your tasks.
5. Evaluate how well you are performing on the job.

6. Evaluate yourself in terms of how well you handle difficulties that arise on the
job.

Note: The rating scale ranged from (1) = veg infrmuently to (5) = vgg fluently.

178
APPENDIX I

Self-Efficacy Items

1.

2.

I am prepared to function effectively on this job because of my past experience.

Previous experience has taught me that I can meet my current work
responsibilities.

My past experiences and accomplishments increase my confidence that I can
perform successfully in this organization.

Prior training and experience gives me assurance that I can accomplish my work
goals.

I believe that I will be successful in my new position based on past experience
and performance.

I feel confident in my skills and ability to perform my job duties based on past
experiences.

Note: Items5 and 6 were developed and added to the original Major (1990) 4—item
scale.

The rating scale ranged from (1) = trong1y disagrg to (5) = trongly agr_’ee.

179
APPENDD( J

Interdgpgndence Items

To what extent:

1.

2.

Must your job activities be coordinated with those of your workgroup?

Do you depend on others to complete their work before you can get your own
work done?

DO you have to work with your workgroup to get your job done?

Do the tasks you perform require you to check with or collaborate with others in
your workgroup?

Do other workgroup members depend on your work in order to complete their
own tasks?

Do you have to talk to other workgroup members to get your job done?
Is the work you do a result of the combined efforts of several individuals?

Do the tasks you perform require you to work with your coworkers?

Note: Items 2, 6, and 8 were developed and added to the original Major (1990) five-
item scale.

Respondents answered these items on a five-point scale ranging from (1) = eg little
extent to (5) = eg geat extent.

1 80
APPENDIX K

Complexity Items

Exceptions

TO what extent:

1. Are your tasks the same from day-to-day?

2. Do you do about the same job in the same way most of the time?
3. Do you basically perform repetitive activities in doing your job?
4. Would you say your work is routine?

5. Are your duties repetitious?

Analyzabilig

To what extent:

1.

Is there an understandable sequence of steps that can be followed in carrying out
your work?

Is there a clearly defined body Of knowledge of subject matter which can guide
you in doing your work?

Is there an understandable sequence of steps that can be followed in doing your
work?

Can you actually rely on established procedures and practices to do your work?

Is there a clearly known way to do the major types of work you normally
encounter?

Note. Respondents answered these items on a five-point scale ranging from (1) = ygy
little extent to (5) = m ggeat extent

18 l
APPENDD( L

Discretion Items

To what extent:

1.

2.

Does your job provide you freedom to act independently of your boss?

Do you have the freedom to choose the order in which different parts of your job
are done?

Does your job allow you to set your own work targets/objectives?

Can you choose the pace at which you work?

Can you choose whom you deal with in order to carry out your job duties?‘
Are you allowed to decide how to schedule your work?

Are you able to choose the methods you use to do your work?

Note. Items 4, 6, and 7 were developed and added to the scale.

‘Item deleted from final discretion scale based on factor analysis results.

Respondents answered these items on a five-point scale ranging from (1) = eg little
extent to (5) = §_ry ggeat extent.

182
APPENDIX M

Sum’sor Functional Information-Sharing Items

How frequently:

1.

2.

Does your supervisor give you information on how to perform your job?
Does your supervisor tell you how to complete your job duties?

Does your supervisor share information with you on how to deal with problems
on the job?‘

Does your supervisor give you information on how to meet your job
responsibilities?

Does your supervisor explain to you how to use the equipment necessary to
perform your job?

Does your supervisor describe to you how to perform your tasks?

Note. ‘This item was deleted from the final supervisor functional information-sharing
scale based on factor analysis results.

Respondents answered these items on a five-point scale ranging from (1) = veg
infiguently to (5) = eg figuently.

1 83
APPENDIX N

Sm’sor Evaluative Information-Sharing Items

How frequently:

1. Does your supervisor indicate how well you are meeting your job
responsibilities?

2. Does your supervisor share with you how well you are performing your tasks?

3. Does your supervisor tell you how well you are succeeding in meeting your work
goals?

4. Does your supervisor tell you how well you are doing on the job?

5. Does your supervisor inform you about how well you are handling difficulties on
your job?

6. Does your supervisor tell you how well you are performing your assignments?

Note. Respondents answered these items on a five-point scale ranging from (1) = veg
infiguently to (5) = eg figuently.

1 84
APPENDD( O

Coworker Functional Information-Sharing Items

How frequently:

1. Do your coworkers give you information on how to perform your job?

2. Do your coworkers tell you how to complete your job duties?

3. Do your coworkers share information with you on how to deal with problems on
the job?

4. Do your coworkers give you information on how to meet your job
responsibilities?

5. Do your coworkers tell you how to use the equipment necessary to perform your
job?

6. DO your coworkers describe to you how to perform your tasks?

Note. Respondents answered these items on a five-point scale ranging from (1) = veg
M uently to (5) = _eg figuently.

1 85
APPENDIX P

Coworker Evaluative Information-Sharing Items

How frequently:

1.

2.

Do your coworkers indicate how well you are meeting your job responsibilities?
Do your coworkers share with you how well you are performing your tasks?

Do your coworkers tell you how well you are succeeding in meeting your work
goals?

Do your coworkers tell you how well you are doing on the job?

Do your coworkers inform you about how well you are handling difficulties on
yourjob?

Do your coworkers tell you how well you are performing your assignments?

Note. Respondents answered these items on a five-point scale ranging from (1) = m
M uently to (5) = eg figuently.

l 86
APPENDIX Q

Leader-Member Exchange Items

1.

DO you usually feel that you know where you stand or do you usually know how
satisfied your immediate supervisor is with what you do?
(4) Always know where I stand

(3) : Usually know where I stand
(2) _ Seldom know where I stand
(1) _ Never know where I stand

How well do you feel that your immediate supervisor understands your problems
and needs?

(4) _ Completely

(3) _ Well enough

(2) _ Some, but not enough
(1) _ Not at all

How well do you feel that your immediate supervisor recognizes your potential?
(4) Fully

(3) : As much as the next person
(2) _ Some, but not enough
(1) _ Not at all

Regardless of how much formal authority your immediate supervisor has built
into his or her position, what are the chances that he or she would be personally
inclined to use power to help you solve problems in your work?

(4) _ Certainly would
(3) _ Probably would
(2) _ Might or might not
(1) _ No chance

Again, regardless of the amount of formal authority your immediate supervisor
has, to what extent can you count on him or her to "bail you out" at his or her
expense when you really need it?

(4) Certainly would

(3) Probably would

(2) Might or might not

(1) No chance

187
APPENDIX Q (cont’d.)

6. I have enough confidence in my immediate supervisor that I would defend and
justify his or her decisions if he or she were not present to do so.

(4) _ Certainly would
(3) _ Probably would
(2) _ Maybe

(I) _ Probably not

7. How would you characterize your relationship with your immediate supervisor?
(4) Extremely effective

(3) : Better than average
(2) _ About average
(1) _ Less than average

l 88
APPENDIX R

Team-Member Exchange Items

1. How often do you make suggestions about better work methods to coworkers?

(4) _ Almost always
(3) __ Fairly often
(2) _ Seldom

(1) _ Almost never

2. Do other members of your workgroup usually let you know when you do
something that makes their jobs easier (or harder)?

(4) _ Almost always
(3) _ Fairly often
(2) _ Seldom

(l) _ Almost never

3. How often do you let coworkers know when they have done something that
makes your job easier (or harder)?

(4) _ Almost always
(3) _ Fairly often
(2) _ Seldom
(l) _ Almost never
4. How well do other members of your workgroup recognize your potential?
(5) _ Fully
(4) _ Quite a bit
(3) _ Some
(2) _ A little
(1) _ Not at all

5. How well do other members of your workgroup understand your problems and
needs?

(4) _ Not at all

(3) _ Some but not enough
(2) __ Well enough

(1) _ Completely

189
APPENDDI R (cont’d.)

6. How flexible are you about switching job responsibilities to make things easier
for your coworkers?
(4) I’m always willing to switch

(3) _ I’m usually willing to switch
(2) _ I’m reluctant, but will switch when I really have to
(1) _ I’m not willing to switch
7. In busy situations, how often do your coworkers ask you to help out?
(4) _ Almost always
(3) _ Usually
(2) _ Seldom
(1) _ Almost never

8. In busy situations, how often do you volunteer your efforts to help others in your
workgroup?

(4) _ Almost always
(3) _ Usually
(2) _ Seldom
(l) _ Almost never
9. How willing are you to help finish work that had been assigned to others?
(4) _ Not willing -- they should finish their own work
(3) _ Might or might not
(2) _ Probably would
(1) __ Certainly would

10. How willing are other members of your workgroup to help finish work that was
assigned to you?

(4) _ Not willing

(3) _ Might or might not
(2) _ Probably would
(1) _ Certainly would

Note. This scale was deleted from the data analyses based on factor analysis results.

190
APPENDIX S

Masteg Items
COWORKER SURVEY

l. The newcomer has learned how to perform all the tasks required by the job.‘

2. The newcomer is competent in carrying out his/her job assignments.

3. The new employee completes tasks in a timely manner.

4. The newcomer has developed a good "system" for performing his/her duties.

5. The newcomer has acquired sufficient knowledge to perform duties without help.‘
6. The newcomer has mastered the "ins and outs" of the job.‘

7. The newcomer is achieving what is expected of him/her on the job.

8. The new employee is successfully meeting his/her job responsibilities.
NEWCOMER SURVEY

1. I have learned how to perform all the tasks required by my job.‘

2. I am competent in carrying out my job assignments.

3. I complete my tasks in a timely manner.

4. I have developed a good "system" for performing my duties.

5. I have acquired sufficient knowledge to perform my duties without help.‘

6. I have mastered the "ins and outs" of my job.‘

7. I am achieving what is expected of me on the job.

8. I am successfully meeting my job responsibilities.

15.0193 ‘Items were deleted from the final mastery scale based on factor analysis results.

Rating scale ranged from (1) = trongly disaw to (7) = trongly am.

191
APPENDIX T

Innovation Items

COWORKER SURVEY

1.

The new employee has made an attempt to redefine his/her role and change what
he/she is required to do.

2. The newcomer has changed the mission or purpose of his/her role.

3. The new employee performs the job according to normal procedures.‘

4. The newcomer has tried to change the procedures for doing his/her job and to institute
new work goals.

5. The new employee has added new responsibilities to his/her role.

6. The newcomer has developed new methods for getting work accomplished.

7. The newcomer has attempted to change the tasks he/she is required to do as part of
his/her job.

NEWCOMER SURVEY

1. I have made an attempt to redefine my role and change what I am required to do.

2. I have changed the mission or purpose of my role.

3. The procedures for performing my job are generally appropriate in my view.‘

4. I have tried to change the procedures for doing my job and to institute new work
goals.

5. I have added new responsibilities to my role.

6. I have developed new methods for getting my work accomplished.

7. I have attempted to change the tasks I am required to do as part of my job.

Note: ‘Item deleted from final innovation scale based on factor analysis results.

Rating scale ranged from (1) = not at all to (5) = o a ggpat extent.

 

192
APPENDIX U

Workgpup Influence Items

NEWCOMER SURVEY

1. My supervisor uses my advice.

2. My coworkers or supervisor ask me for advice.

3. My work group includes me in decisions about our work.

4. My coworkers act on my suggestions.

5. My supervisor accepts my suggestions about work.

6. My coworkers accept my ideas for how work can be done better.
7. My supervisor asks me for input on decisions.

8. My supervisor listens to my ideas for improving my job.

COWORKER SURVEY

1. The supervisor uses advice from the newcomer.

2. Coworkers ask the new employee for advice.

3. The work group includes the newcomer in decisions about work.

4. The work group acts on the newcomers suggestions.

5. The supervisor accepts the new employee’s suggestions about work.

6. Coworkers accept the newcomer’s ideas for how work can be done better.
7. The supervisor asks the new employee for input on decisions.

8. The supervisor listens to the newcomer’s ideas for improving his/her job.

M21 Rating scale ranged from (1) = strongly disgm to (7) = strongly gm.

193
APPENDIX V

Study Analyses

ANALYSIS 1 - 3 (Stratggy) x 2 (Content) x 3 (Time on the Job) ANOVA

Hypothesis 1. A three-way interaction between information content (functional,
evaluative), information-seeking strategy, and time on the job will affect information-
seeking.

Hymthesis 2. Newcomers will use experimentation more frequently than inquiry of
coworkers and inquiry of supervisors to acquire functional information.

Hypothesis 3. Newcomers will use observation more fiequently than inquiry of coworkers
and inquiry of supervisors to acquire functional information.

Hypothesis 4. Newcomers will inquire of coworkers for functional information to a
greater extent than inquiring of supervisors for functional information.

Hypothesis 5. Newcomers will use experimentation more frequently than inquiry of
supervisors and inquiry of coworkers to acquire evaluative information.

Hypothesis 6. Newcomers will use observation more frequently than inquiry of
supervisors and inquiry of coworkers to acquire evaluative information.

Hyppthesis 7. Newcomers will inquire of supervisors for evaluative information to a
greater extent than inquiring of coworkers for evaluative information.

Hypothesis 8. Inquiry of supervisors for functional information will decrease with greater
time on the job.

Hypothesis 9. Inquiry of coworkers for functional information will decrease with greater
time on the job.

H thesis 10. Inquiry of supervisors for evaluative information will increase with greater
time on the job.

Hymthesis 11. Inquiry of coworkers for evaluative information will increase with greater
time on the job.

194
APPENDIX V (cont’d.)

ANALYSIS 2 - MODERATED HIERARCHICAL REGRESSION

Hypothesis 12. Task interdependence will be positively related to inquiry of supervisors
for functional information.

Hypothesis 12a. This relationship will be moderated by the newcomer’s self-efficacy.
For low self-efficacy individuals, a strong, positive relationship between task
interdependence and inquiry of supervisors for functional information is expected; for
high self-efficacy individuals this relationship should be weaker.

Hypothesis 20. Exceptions (variety) will be positively related to inquiry of supervisors for
functional information.

Hypothesis 20a. Self-efficacy will moderate this relationship. Exceptions will be
positively related to inquiry of supervisors for functional information for individuals
low on self-efficacy. For individuals high on self-efficacy, the relationship should be
weaker.

Hypothesis 28. Analyzability will be positively related to inquiry of supervisors for
functional information.

Hypothesis 28a. Self—efficacy will moderate this relationship. For low self-efficacy
individuals, analyzability will be negatively related to inquiry of supervisors for
functional information. For high self-efficacy individuals, the relationship will be
weaker.

Hymthesis 36. Discretion will be negatively related to inquiry of supervisors for
functional information.

Hypothesis 36a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, discretion will be positively related to inquiry of supervisors for
functional information. For high self-efficacy individuals, the relationship will be
negative.

ANALYSIS 3 - MODERATED HIERARCHICAL REGRESSION

Hypothesis 13. Task interdependence will be positively related to inquiry of coworkers
for functional information.

Hypothesis 13a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, a strong, positive relationship between task interdependence and inquiry
of coworkers for functional information is expected; for high self-efficacy individuals
this relationship should be weaker.

195
APPENDIX V (cont’d.)

Hymthesis 21 . Exceptions (variety) will be positively related to inquiry of coworkers for
functional information.

Hyppthesis 21a. Self-efficacy will moderate this relationship. Exceptions will be
positively related to inquiry of coworkers for functional information for individuals
low on self-efficacy. For individuals high on self-efficacy, the relationship should be
weaker.

Hymthesis 29. Analyzability will be negatively related to inquiry of coworkers to a
greater extent for functional information.

Hypothesis 29a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, analyzability will be negatively related to inquiry of coworkers for
functional information. For high self-efficacy individuals, the relationship will be
weaker.

Hymthesis 37. Discretion will be negatively related to inquiry of coworkers for functional
information.

Hypothesis 37a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, discretion will be positively related to inquiry of coworkers for functional
information. For high self—efficacy individuals, the relationship will be negative.

ANALYSIS 4 - MODERATED HIERARCHICAL REGRESSION

Hypothesis 14. Task interdependence will be negatively related to observation for
functional information.

Hypothesis 14a. Newcomer self-efficacy will moderate this relationship. For
individuals with low self-efficacy, there should be a negative relationship between
task interdependence and observation for functional information. For high self-
efficacy individuals, task interdependence should have no relationship to observation
for functional information.

Hypothesis 22. Exceptions (task variety) will be negatively related to observation for
functional information.

Hypothesis 22a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, exceptions should be negatively related to observation to acquire
functional information. For high self-efficacy individuals, exceptions should be
positively related to observation for functional information.

196
APPENDIX V (cont’d.)

Hypothesis 30. Analyzability will be positively related to observation for frmctional
information.

Hypothesis 30a. Self-efficacy will moderate this relationship. Analyzability will be
positively related to observation for functional information for low self-efficacy
individuals. The relationship will be negative for high self-efficacy individuals.

Hypothesis 38. Discretion will be positively related to observation for functional
information.

Hypothesis 38a. Self-efficacy will moderate this relationship. Discretion will be
negatively related to observation for functional information for low self—efficacy
individuals. For high self-efficacy individuals, the relationship will be positive.

ANALYSIS 5 - MODERATED I-IIERARCHICAL REGRESSION

Hypothesis 15. Task interdependence will be negatively related to experimentation for
functional information.

Hypothesis 15a. Self-efficacy will moderate this relationship. For individuals with
low self-efficacy, there should be a negative relationship between task
interdependence and experimentation for functional information. For high self-
efi‘icacy individuals, task interdependence should have no relationship to
experimentation for functional information.

Hymthesis 23. Exceptions will be negatively related to experimentation for functional
information.

Hymthesis 23a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, exceptions should be negatively related to experimentation to acquire
functional information. For high self-efficacy individuals, exceptions should be
positively related to experimentation for functional information.

Hypothesis 31. Analyzability will be positively related to experimentation for functional
information.

Hypothesis 31a. Self-efficacy will moderate this relationship. Analyzability will be
positively related to experimentation for functional information for low self-efficacy
individuals. The relationship will be negative for high self-efficacy individuals.

197
APPENDDI V (cont’d.)

Hypothesis 39. Discretion will be positively related to experimentation for functional
information.

Hypothesis 39a. Self-efficacy will moderate this relationship. Discretion will be
negatively related to experimentation for functional information for low self-efficacy
individuals. For high self-efficacy individuals, the relationship will be positive.

ANALYSIS 6 - MODERATED HIERARCHICAL REGRESSION

Hypothesis 16. Task interdependence will be positively related to inquiry of supervisors
for evaluative information.

Hymthesis 16a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, task interdependence will be negatively related to inquiry of supervisors
for evaluative information. For high self-efficacy individuals, task interdependence
will be positively related to inquiry of supervisors for evaluative information.

Hypothesis 24. Exceptions (variety) will be positively related to inquiry of supervisors for
evaluative information.

Hymthesis 24a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, exceptions will be negatively related to inquiry of supervisors for
evaluative information. For high self-efficacy individuals, exceptions will be
positively related to inquiry of supervisors for evaluative information.

Hypothesis 32. Analyzability will be negatively related to inquiry of supervisors for
evaluative information.

Hyppthesis 32a. Self-efficacy will moderate this relationship. For individuals low on
self-efficacy, analyzability will be positively related to inquiry of supervisors for
evaluative information. For high self-efficacy individuals, the relationship will be
negative.

Hymthesis 40. Discretion will be positively related to inquiry of supervisors for
evaluative information.

Hymthesis 40a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, the relationship between discretion and inquiry of supervisors for
evaluative information will be negative. For high self-efficacy individuals, the
relationship will be positive.

198
APPENDDI V (cont’d.)

ANALYSIS 7 - MODERATED HIERARCHICAL REGRESSION

Hypothesis 17. Task interdependence will be positively related to inquiry of coworkers
for evaluative information.

Hypothesis 17a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, task interdependence will be negatively related to inquiry of coworkers
for evaluative information. For high self-efficacy individuals, task interdependence
will be positively related to inquiry of coworkers for evaluative information.

Hymthesis 25. Exceptions (variety) will be positively related to inquiry of coworkers for
evaluative information.

Hymthesis 25a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, exceptions will be negatively related to inquiry of coworkers for
evaluative information. For high self-efficacy individuals, exceptions will be
positively related to inquiry of coworkers for evaluative information.

Hyppthesis 33. Analyzability will be negatively related to inquiry of coworkers for
evaluative information.

Hypothesis 33a. Self-efficacy will moderate this relationship. For individuals low on
self-efficacy, analyzability will be positively related to inquiry of coworkers for
evaluative information. For high self-efficacy individuals, the relationship will be
negative.

Hypothesis 41. Discretion will be positively related to inquiry of coworkers for evaluative
information.

Hypothesis 41a. Self—efficacy will moderate this relationship. For low self-efficacy
individuals, the relationship between discretion and inquiry of coworkers for
evaluative information will be negative. For high self-efficacy individuals, the
relationship will be positive.

ANALYSIS 8 - MODERATED HIERARCHICAL REGRESSION

 

Hymthesis 18. Task interdependence will be negatively related to observation for
evaluative information.

Hypothesis l8a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, interdependence should have no relationship to observation for evaluative
information. For individuals high on self-efficacy, there should be a negative
relationship between interdependence and observation for evaluative information.

199
APPENDIX V (cont’d.)

Hymthesis 26. Exceptions will be negatively related to observation for evaluative
information.

Hypothesis 26a. Self-efficacy will moderate this relationship. Exceptions will be
negatively related to observation for evaluative information for low self-efficacy
individuals. Exceptions will be positively related to observation for evaluative
information for high self-efficacy individuals.

Hymthesis 34. Analyzability will be positively related to observation for evaluative
information.

Hypothesis 34a. Self-efficacy will moderate this relationship. Analyzability will be
positively related to observation for evaluative information for low self-efficacy
individuals. For high self-efficacy individuals, the relationship will be negative.

Hymthesis 42. Discretion will be positively related to Observation for evaluative
information.

Hymthesis 42a. Self-efficacy will moderate this relationship. For high self-efficacy
individuals, discretion will be positively related to observation for evaluative
information. For low self-efficacy individuals, the relationship will be weaker.

ANALYSIS 9 - MODERATED HIERARCHICAL REGRESSION

Hymthesis 19. Task interdependence will be negatively related to experimentation for
evaluative information.

Hypothesis 19a. Self-efficacy will moderate this relationship. For low self-efficacy
individuals, interdependence should have no relationship to experimentation for
evaluative information. For individuals high on self-efficacy, there should be a
negative relationship between interdependence and experimentation for evaluative
information.

Hyppthesis 27. Exceptions will be negatively related to experimentation for evaluative
information.

Hypothesis 27a. Self-efficacy will moderate this relationship. Exceptions will be
negatively related to experimentation for evaluative information for low self-efficacy
individuals. Exceptions will be positively related to experimentation for evaluative
information for high self-efficacy individuals.

200
APPENDIX V (cont’d.)

Hymthesis 35 . Analyzability will be positively related to experimentation for evaluative
information.

Hymthesis 35a. Self-efficacy will moderate this relationship. Analyzability will be
positively related to experimentation for evaluative information for low self-efficacy
individuals. For high self-efficacy individuals, the relationship will be negative.

Hymthesis 43. Discretion will be positively related to experimentation for evaluative
information.

Hypothesis 43a. Self-efficacy will moderate this relationship. For high self-efficacy
individuals, discretion will be positively related to experimentation for evaluative
information. For low self-efficacy individuals, the relationship will be weaker.

ANALYSIS 10 - REGRESSION

Hymthesis 44. Sumrvisor sharing and coworker sharing of functional information will be
positively related to the extent to which the newcomer seeks functional information
through inquiry of supervisors.

Hypothesis 52. Individuals who report a higher quality LMX relationship will be more
likely to inquire of supervisors for functional information.

ANALYSIS ll - REGRESSION

Hypothesis 45 . Sumrvisor sharing and coworker sharing of functional information will be
positively related to the extent to which the newcomer seeks functional information
through inquiry of coworkers.

Hymthesis 54. Individuals who report a higher quality TMX relationship will be more
likely to inquire of coworkers for functional information.

ANALYSIS 12 - REGRESSION

Hymthesis 46. Sumrvisor sharing and coworker sharing of frmctional information will be

positively related to the extent to which the newcomer seeks functional information
through observation.

201
APPENDIX V (cont’d.)

ANALYSIS 13 - REGRESSION

Hypothesis 4?. Sumrvisor sharing and coworker sharing of functional information will be
positively related to the extent to which the newcomer seeks functional information
through experimentation.

ANALYSIS l4 - REGRESSION

Hymthesis 48. Sugrvisor sharing and coworker sharing of evaluative information will
be positively related to the extent to which the newcomer seeks evaluative information
through inquiry of supervisors.

Hymthesis 53. Individuals who report a higher quality LMX relationship will be more
likely to inquire of supervisors for evaluative information.

ANALYSIS 15 - REGRESSION

Hypothesis 49. Sugrvisor sharing and coworker sharing of evaluative information will

be positively related to the extent to which the newcomer seeks evaluative information
through inquiry of coworkers.

 

Hypothesis 55. Individuals who report a higher quality TMX relationship will be more
likely to inquire of coworkers for evaluative information.

ANALYSIS l6 - REGRESSION

Hyppthesis 50. Sugrvisor sharing and coworker sharing of evaluative information will be
positively related to the extent to which the newcomer seeks evaluative information
through observation.

 

ANALYSIS l7 - REGRESSION

Hypothesis 51. Supprvisor sharing and coworker sharing of evaluative information will be
positively related to the extent to which the newcomer seeks evaluative information
through experimentation.

202
APPENDIX V (cont’d.)

ANALYSIS 18 - REGRESSION

Hyppthesis 56. Individuals who seek functional information to a greater extent (across all
strategies) will exhibit greater mastery of the job.

Hypothesis 57. Time on the job will moderate the relationship between inquiry of
supervisors for functional information and mastery. With little time on the job, inquiry of
supervisors for functional information should be positively related to mastery. With
greater time on the job, inquiry of supervisors for functional information should be
negatively related to mastery.

Hypothesis 58. Time on the job will moderate the relationship between inquiry of
coworkers for functional information and mastery. With little time on the job, inquiry of
coworkers for functional information should be positively related to mastery. With greater
time on the job, inquiry of coworkers for functional information should be negatively
related to mastery.

ANALYSIS l9 - REGRESSION

Hypothesis 59. Individuals who seek evaluative information to a greater extent (across all
strategies) will exhibit greater mastery of the job.

ANALYSIS 20 - HIERARCHICAL REGRESSION

Hyppthesis 60. The use of observation and experimentation above and beyond the use of
inquiry of supervisors and coworkers to acquire functional information should be
positively related to innovation; the use of inquiry of supervisors and coworkers beyond
observation and experimentation to acquire functional information should not have a
unique impact on innovation.

ANALYSIS 21 - REGRESSION

Hypothesis 61. Newcomers who seek evaluative information to a greater extent (across all
strategies) will exhibit greater levels of innovation on the job.

203
APPENDDI V (cont’d.)

ANALYSIS 22 - REGRESSION
Hypothesis 62. Mastery will be positively related to workgroup influence.

Hypothesis 63. Mastery will moderate the relationship between innovation and workgroup
influence. Innovation will be positively related to workgroup influence for individuals
high on task mastery; innovation should have no relationship to workgroup influence for
individuals low on task mastery.

204
APPENDIX W

Principle Axis Factor Matrix for Observation

and Coworker Sharing of Functional Information

 

 

Factor

1 2
Cowork Func6 ._89_ .29
Cowork Funcl 19_ .29
Cowork Func4 19 .20
Cowork Func2 15 .30
Cowork Func3 ,Q .24
Cowork Func5 ,65 .27
Observe Func5 .21 _8_3_
Observe Func3 .28 7_9
Observe Func2 .24 ._7_2
Observe Func6 .24 J_()_
Observe Func4 .30 66
Observe Funcl .30 __6_5_

EIGEN VALUE 6.4 1.7

LIST OF REFERENCES

LIST OF REFERENCES

Allen, N. J., & Meyer, J. P. (1990). Organizational socialization tactics: A
longitudinal analysis of links to newcomers’ commitment and role orientation.
Academy of Management Journal, 3_3_, 847-858.

Arps, G. F. (1917). A preliminary report on work with knowledge versus work
without knowledge of results. Psychological Review, _24, 449-455.

Ashford, S. J. (1986). Feedback-seeking in individual adaptation: A resource
perspective. Academy of Management Journal, 22, 465-487.

Ashford, S. J., & Cummings, L. L. (1983). Feedback as an individual resource:
Personal strategies of creating information. Organizational Behavior and Human
Performance, 2, 370-398.

Ashford, S. J., & Cummings, L. L. (1985). Proactive feedback seeking: The
instrumental use of the information environment. Journal of Occupational
Psychology, 58, 67-79.

Ashford, S. J ., & Taylor, M. S. (1990). Adaptation to work transitions: An
integrative approach. Research in Personnel and Human Resources Managemeng,
8 1-39.
—’

Baldwin, T. T., & Ford, J. K. (1988). Transfer of training: A review and directions
for future research. Personnel Psychology, 4_1_, 63-105.

Bandura, A. (1977a). Self-efficacy: Toward a unifying theory of behavioral change.
Psychological Review, _84, 191-215.

Bandura, A. (1977b). Social learning theog. Englewood Cliffs, NJ: Prentice-Hall.

Bandura, A. (1982). Self-efficacy mechanism in human agency. American
Psycholog: g, 3_7_, 122-147.

Banick, M. R., & Mount, M. K. (1991). The big five personality dimensions and job
performance: A meta-analysis. Personnel Psychology, 4_4_, 1-26.

205

206

Becker, T. E., & Klirnoski, R. J. (1989). A field study of the relationship between the
organizational feedback environment and performance. Personnel Psychology, fl,
343-358.

Brett, J. M. (1984). Job transitions and personal and role development. Research in
Personnel and Human Resources Managemeng, _2_, 155-185.

Burke, R. J., & Bolf, C. (1986). Learning within organizations: Sources and content.
Psychological Reports, 3, 1187-1196.

Campbell, D. (1988). Task complexity: A review and analysis. Academy of
Management Review, 1_3_, 40-52.

Daft, R. L., & Macintosh, N. B. (1981). A tentative exploration into the amount and
equivocality of information processing in organizational work units.
Administrative Science Quarterly, .2_6, 207-224.

Dansereau, F., Jr., Graen, G., & Haga, W. J. (1975). A vertical dyad linkage approach
to leadership within formal organizations. Organizational Behavior and Human
Performance, 1_3, 46-78.

Dewhirst, H. D. (1971). Influence of perceived information-sharing norms on

communication channel utilization. Academy of Management Journal, _14, 305-
315.

Eden, D., & Shani, A. B. (1982). Pygmalion goes to boot camp: Expectancy,
leadership, and trainee performance. Journal of Applied Psychology, 61, 194—199.

Evan, W. M. (1963). Peer group interaction and organization socialization: A study of
employee turnover. American Sociological Review, gs, 436-440.

Falcione, R. L., & Wilson, C. E. (1988). Socialization processes in organizations. In
G. M. Goldhaber & G. A. Barnett (Eds), Handbook of organizational
communication (pp. 151-169). Norwood, NJ: Ablex Publishing Corporation.

Feldman, D. C. (1976). A contingency theory of socialization. Administrative
Science Quarterly, A, 433-452.

Feldman, D. C. (1977). The role of initiation activities in socialization. Human
Relations, 39, 977-990.

Feldman, D. C. (1981). The multiple socialization of organization members.
Academy of Management Review, 6, 309-318.

207

Feldman, D. C. (1989). Socialization, resocialization, and training: Reframing the
research agenda. In LL. Goldstein (Ed.), Training and development in
organizations (pp. 376-416). San Francisco: Jossey-Bass.

Feldman, D. C., & Brett, J. M. (1983). Coping with new jobs: A comparative study
of new hires and job changers. Academy of Management Journal, 16, 258-272.

Fisher, C. D. (1982). Idenm ' g the ogtgomes of socialization: Two studies. (Tech.
Rep. No. ONR-8) College Station, TX: Texas A & M University, College of
Business Administration.

Fisher, C. D. (1985). Social support and adjustment to work: A longitudinal study.
Journal of Management, fl, 39-53.

Fisher, C. D. (1986). Organizational socialization: An integrative review. In K.
Roland & G. Ferris (Eds), Research in momel and human resource
management (Vol. 4, pp. 101-145). Greenwich, CT: JAI Press.

Fried, Y., & Penis, G. R. (1987). The validity of the job characteristics model: A
review and meta-analysis. Personnel Psychology, 46, 287-322.

Gerstberger, P. G., & Allen, T. J. (1968). Criteria used by research and deveIOpment

engineers in the selection of an information source. Journal of Applied
Psychology, 2, 272-279.

Gist, ME. (1987). Self-efficacy: Implications for organizational behavior and human
resource management. Academy of Management Review. Q, 472-485.

Gist, M.E., Schwoerer, C., & Rosen, B. (1989). Effects of alternative training
methods on self-efficacy and performance in computer software training. Journal
of Applied Psychology. 11, 884-891.

Goldstein, A. P., & Sorcher, M. (1974). Changing sumrvisor behavior. New York:
Pergarnon Press.

Graen, G. (1976). Role making processes within complex organizations. In M. D.

Dunnette (Ed.), The hgndbogkLOf industrial and organizational psychology.
Chicago: Rand McNally.

Graen, G., Cashman, J. F., Ginsburg, S., & Schiemann, W. (1977). Effects of linking-
pin quality on the quality of working life of lower participants. Administrative
Science Quarterly, _21, 491-504.

208

Graen, G., & Ginsburgh, S. (1977). Job resignation as a function of role orientation
and leader acceptance: A longitudinal investigation of organizational assimilation.

Organizational Behavior and Human Performance, 1_9_, 1-17.

Graen, G., Novak, M. A., & Sommerkamp, P. (1982). The effects of leader-member
exchange and job design on productivity and satisfaction: Testing a dual

attachment model. Organizational Behavior and Human Performance, x, 109-
131.

Graen, G., Orris, J.B., & Johnson, T. (1973). Role assimilation processes in a
complex organization. Journal of Vocational Behavior _3_, 395-420.

 

Graen, G., & Scandura, T. A. (1987). Toward a psychology of dyadic organizing.
Resegrch in Organizational Behavior, 2, 175-208.

Greller, M. M., & Herold, D. M. (1975). Sources of feedback: A preliminary
investigation. Organizational Behavior and Human Performance, g, 244—256.

Hackrnan, J. R., & Oldham, G. R. (1980). Work redesign. Reading, MA: Addison-
Wesley.

Hage, J., Aiken, M., & Marrett, C. B. (1971). Organization structure and
communications. American Sociological Review, 3_6_, 860-871.

Hanser, L. M., & Muchinsky, P. M. (1978). Work as an information environment.
Organizational Behavior and Human Performance, 3;, 47-60.

Heslin, R., Blake, B., & Rotton, J. (1972). Information search as a function of
stimulus uncertainty and the importance of the response. Journal of Personalig
and Social Psychology, Q, 333-339.

Hughes, B. C. (1958). Men and their work. Glencoe, IL: Free Press.

Ilgen, D. R, Fisher, C. D., & Taylor, M. S. (1979). Consequences of individual

feedback on behavior in organizations. Journal of Applied Psychology, fi, 349-
37 1.

Jablin, F. M. (1982). Organizational communication: An assimilation approach. In

M. E. Roloff & C. R. Berger (Eds), Social cognition and communication (pp.
255-286). Beverly Hills, CA: Sage.

Jablin, F. M. (1984). Assirnilating new members into organizations. In R. N.

Bostrom (Ed.), Communication yearbook 8 (pp. 594-626). Beverly Hills, CA:
Sage.

209

Jablin, F. M. (1985). Task/work relationships: A life-span perspective. In M. L.

Knapp & G. R. Miller (Eds), Handbook of intemrsonal communication (pp. 615-
654). Beverly Hills, CA: Sage.

Jablin, F. M., & Sussman, L. (1983). Organizational group communication: A review
of the literature and model of the process. In H. H. Greenbaum, R. L. Falcione &

S. A. Hellweg (Eds), Organizational communication: Abstracts, analysis, and
overview (pp. 11-50). Beverly Hills, CA: Sage.

Jones, G. R. (1983). Psychological orientation and the process of organizational
socialization: An interactionist perspective. Academy of Management Review, 8,
464-474.

Jones, G. R. (1986). Socialization tactics, self-efficacy, and newcomers’ adjustments
to organizations. Academy of Management Journal, 22, 262-279.

Katz, R (1980). Time and work: Towards an integrative perspective. In B. Staw &

L. Cummings, (Eds), Research in organizational behavior (Vol. 2, pp. 81-127).
Greenwich, CT: JAI Press.

Katz, R. (1985). Organizational stress and early socialization experiences. In T. A.

Beehr & R. S. Bhagat (Eds), Human stress and cogjtions in organizations: An
integrated mfive (pp. 117-139). New York: Wiley.

Kirmeyer, S. L., & Lin, T-R. (1987). Social support: Its relationship to observed
communication with peers and superiors. Academy of Management Journal, _3_(_),
138-151.

Kozlowski, S. W. J., & Doherty, M. L. (1989). Integration of climate and leadership:
Examination of a neglected issue. Journal of Applied Psychology, '_7_4_, 546-553.

Kozlowski, S. W. J., Ford, J. K., & Smith, E. M. (1993). Training concepts,
principles, and gpidelines for the aguisition, transfer, and enhancement of tea_rr_r
tactical decision making skills I: A conceptual framework and literature review.
Final Report No. IPPSR/ORG/PSY 93-l.l prepared for the Naval Training
Systems Center, Orlando, FL (Contract No. N61339-91-C-0117).

Kraiger, K., Ford, J. K., & Salas, E. (1993). Application of cognitive, skill-based,
and affective theories of learning outcomes to new methods of training evaluation.
Journal of App1ied Psychology, _7_8_, 311-328.

Larson, J. A., Jr. (1989). The dynamic interplay between employees’ feedback-

seeking strategies and supervisors’ delivery of performance feedback. Academy
of Management Review, 14, 408-422.

210

Locke, E. A., Frederick, E., Lee, C. & Bobko, P. (1984). Effect of self-efficacy,
goals, and task strategies on task performance. Journal of Applied Psychology,
Q, 241-251.

Lott, A. J., & Lott, B. E. (1961). Group cohesiveness, communication level, and
conformity. Journal Of Abnormal and Socjial Psychology, 6;, 408-412.

Lott, A. J ., & Lott, B. E. (1965). Group cohesiveness as interpersonal attraction: A

review of relationships with antecedent and consequent variables. Psychological
Bulletin _64, 259-309.

 

Louis, M. R. (1980). Surprise and sense making: What newcomers experience in
entering unfamiliar organizational settings. Administrative Science Quarterly, 25,
226-251.

Louis, M. R. (1990). Acculturation in the workplace: Newcomers as lay
ethnographers. In B. Schneider (Ed.), Organizational climate and culture (pp. 85-
129). San Francisco: Jossey-Bass.

Louis, M. R., Posner, B. Z., & Powell, G. N. (1983). The availability and helpfulness
of socialization practices. Personnel Psychology, 36, 857-866.

Major, D. A. (1990). Differential effects of interggtjon content and proaction on
organizational socialization outcomes. Unpublished master’s thesis, Michigan
State University, East Lansing, MI.

Major, D. A., & Kozlowski, S. W. J. (1990). Self-efficacy and proaction in the
organizational socialization context. Paper presented at the 50th Annual Meeting
of the Academy of Management, San Francisco, CA.

March, J. G., & Simon, H. A. (1958). Organizations. New York: Wiley.

Miller, V. D., & Jablin, F. M. (1991). Information seeking during organizational entry:
Influences, tactics, and a model of the process. Academy of Management Review,
16, 92-120.

Morrison, E. W. (1991). Newcomer information seeking: A longitudinal study of

types, modes and sources. Academy of Management Best Pamr Proceedings,
225-229.

Monison, E. W. (1992). An egploratog study of information aguisition during
socialization. Paper presented at the 1992 Conference of the Society for Industrial

and Organizational Psychology, Montreal, Quebec.

211

Morrison, E. W. (1993). Longitudinal study of the effects of information seeking on
newcomer socialization. Journal of Applied Psychology, 3, 173—183.

Monison, E. W., & Bies, R. J. (1991). Impression management in the feedback-
seeking process: A literature review and research agenda. Academy of
Management Review, 16, 522-541.

Nicholson, N. (1984). A theory of work role transitions. Administrative Science
Quarterly, _2_9_, 172-191.

Northcraft, G. B., & Ashford, S. J. (1990). The preservation of self in everyday life:
The effects of performance expectations and feedback context on feedback
inquiry. Organizational Behgvior grid Human Decision Processes. 41, 42-64.

Ostroff, C., & Kozlowski, S. W. J. (1992). Organizational socialization as a learning
process: The role of information acquisition. Personnel Psychology, 45, 849-874.

Ostroff, C., & Kozlowski, S. W. J. (1993). The role of mentoring in the information
gathering process of newcomers during early organizational socialization. Journal
of Vocational Behavior 4;, 170-183.

 

Perrow, C. (1967). A fi'amework for the comparative analysis of organizations.
American Sociological Review, .22, 194-208.

Porter, L. W., Lawler, E. E., III, & Hackrnan, J. R. (1975). Behavior in organizations.
New York: McGraw-Hill.

Reichers, A. (1987). An interactionist perspective on newcomer socialization rates.
Academy of Management Review, fl, 278-287.

Scandura, T. A., & Graen, G. B. Moderating effects of initial leader-member
exchange status on the effects of a leadership intervention. Journal of Applied
Psychology, Q, 428-436.

Schein, E. H. (1968). Organizational socialization and the profession of management.
Industrial Management Review, 2, 1-16.

Schneider, B. (1983). Interactional psychology and organizational behavior. Research
in Organizational Behavior, 5, 1-31.

Seers, A. (1989). Team-member exchange quality: A new construct for role-making
research. Organizational Behavior and Human Decision Processes, _4_3_, 118-135.

212

Sheridan, J. E. (1985). A catastrophe model of employee withdrawal leading to low
job performance, high absenteeism, and job turnover during the first year of

employment. Academy of Management Journal, E, 88-109.

Spector, P. E. (1986). Perceived control by employees: A meta-analysis of studies
concerning autonomy and participation at work. Human Relations 32, 1005- 1016.

 

Terborg, J. R., & Miller, H. E. (1978). Motivation, behavior, and performance: A
closer examination of goal setting and monetary incentives. Journal of Applied

Psychology, 65, 29-39.

Thompson, J. D. (1967). Organizations in action. New York: McGraw-Hill.

Tushman, M. L. (1979). Work characteristics and subunit communication structure:
A contingency analysis. Administrative Science Qparterly, at, 82-98.

Van de Ven, A. H., Delbecq, A. L., Koenig, R., Jr. (1976). Determinants of
coordination modes within organizations. American Sociological Review, 41,
322-338.

Van Maanen, J. (1975). Police socialization: A longitudinal examination of job
attitudes in an urban police department. Administrative Science marterly, 26,
207-228.

Van Maanen, J. (1976). Breaking in: Socialization to work. In R. Dubin (Ed.),
Handbook of work, organization, and socieg. Chicago: Rand McNally.

Van Maanen, J. (197 8). People processing: Strategies of organizational socialization.
Organizational Dmamics, _7_, 19-36.

Van Maanen, J., & Schein, E. H. (1979). Toward a theory of organizational
socialization. Reseryrch in organizational behavior (Vol. 1 pp. 209-264).

Wanous, J. (1980). OrganizationaLl enpy: Recruitmeng, selection, and socialization of
newcomers (pp. 21-84). Reading, MA: Addison-Wesley.

Weiss, H. M. (1978). Social learning of work values in organizations. Journal of
Applied Psychology, 66, 711-718.

West, M. A. (1987). Role innovation in the world of work. British Journal of Social
Psychology, 26, 305-315.

West, M. A., & Farr, J. L. (1989). Innovation at work: Psychological perspectives.
Social Behaviour, 4, 15-30.

 

213

West, M. A., Nicholson, N., & Rees, A. (1987). Transitions into newly created jobs.
Journal of Occupational Psychology, Q, 97-113.

Wheeler, S. (1966). The sn'ucture of formally organized socialization settings. In O.
S. Brim, Jr., & S. Wheeler (Eds), Socialization after childhood: Two essa_y_s_ (pp.
51-116). New York: Wiley.

Withey, M., Daft, R. L., & Cooper, W. H. (1983). Measures of Perrow’s work unit
technology: An empirical assessment and a new scale. Academy of Management
Journal 3, 45-63.

 

Wood, R. E. (1986). Task complexity: Definition of the construct. Organizational
Behavior and Human Decision Processes _31, 60-82.

 

Zuckerman, M., Brown, R. H., Fischler, G. L., Fox, G. A., Lathin, D. R., & Minasian,
A. J. (1979). Determinants of information-seeking behavior. Journal of Research

in Personalig, 13, 161-174.

"iiillllllliiiiilii